{
  "best_metric": NaN,
  "best_model_checkpoint": "./phi-sft-outOO/checkpoint-250",
  "epoch": 2.8665931642778393,
  "eval_steps": 125,
  "global_step": 13000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0,
      "loss": 2.24,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 3.0102999566398115e-06,
      "loss": 2.2025,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.771212547196624e-06,
      "loss": 2.1232,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 6.020599913279623e-06,
      "loss": 2.2101,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 6.989700043360187e-06,
      "loss": 2.3343,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 7.781512503836437e-06,
      "loss": 2.227,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 8.450980400142568e-06,
      "loss": 2.0597,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 9.030899869919434e-06,
      "loss": 2.1001,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 9.542425094393249e-06,
      "loss": 1.9086,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 9.999999999999999e-06,
      "loss": 2.1347,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.041392685158225e-05,
      "loss": 2.0306,
      "step": 11
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.0791812460476248e-05,
      "loss": 2.0697,
      "step": 12
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.1139433523068365e-05,
      "loss": 2.0878,
      "step": 13
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.1461280356782378e-05,
      "loss": 2.1483,
      "step": 14
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.1760912590556813e-05,
      "loss": 1.8576,
      "step": 15
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.2041199826559246e-05,
      "loss": 2.1813,
      "step": 16
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.230448921378274e-05,
      "loss": 1.8889,
      "step": 17
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.2552725051033058e-05,
      "loss": 1.8875,
      "step": 18
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.2787536009528288e-05,
      "loss": 2.1339,
      "step": 19
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.301029995663981e-05,
      "loss": 2.0742,
      "step": 20
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.3222192947339192e-05,
      "loss": 1.8415,
      "step": 21
    },
    {
      "epoch": 0.0,
      "learning_rate": 1.3424226808222062e-05,
      "loss": 1.7386,
      "step": 22
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.3617278360175927e-05,
      "loss": 1.7894,
      "step": 23
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.380211241711606e-05,
      "loss": 1.7763,
      "step": 24
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.3979400086720374e-05,
      "loss": 1.9887,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.4149733479708178e-05,
      "loss": 2.0224,
      "step": 26
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.4313637641589872e-05,
      "loss": 2.1302,
      "step": 27
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.4471580313422191e-05,
      "loss": 1.7324,
      "step": 28
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.4623979978989559e-05,
      "loss": 1.8852,
      "step": 29
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.4771212547196623e-05,
      "loss": 1.9105,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.4913616938342726e-05,
      "loss": 2.0049,
      "step": 31
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5051499783199059e-05,
      "loss": 1.8416,
      "step": 32
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5185139398778874e-05,
      "loss": 1.6674,
      "step": 33
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.531478917042255e-05,
      "loss": 1.814,
      "step": 34
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5440680443502753e-05,
      "loss": 1.7162,
      "step": 35
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5563025007672873e-05,
      "loss": 1.8088,
      "step": 36
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.5682017240669948e-05,
      "loss": 1.8689,
      "step": 37
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.57978359661681e-05,
      "loss": 1.8064,
      "step": 38
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.591064607026499e-05,
      "loss": 1.8098,
      "step": 39
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.6020599913279622e-05,
      "loss": 1.8754,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.6127838567197353e-05,
      "loss": 1.64,
      "step": 41
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.6232492903979005e-05,
      "loss": 1.6888,
      "step": 42
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.6334684555795865e-05,
      "loss": 1.8604,
      "step": 43
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.6434526764861872e-05,
      "loss": 1.9152,
      "step": 44
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.6532125137753435e-05,
      "loss": 1.9345,
      "step": 45
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.662757831681574e-05,
      "loss": 1.7926,
      "step": 46
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.672097857935717e-05,
      "loss": 1.8824,
      "step": 47
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.681241237375587e-05,
      "loss": 1.8944,
      "step": 48
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.6901960800285137e-05,
      "loss": 1.8863,
      "step": 49
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.6989700043360187e-05,
      "loss": 1.6199,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7075701760979363e-05,
      "loss": 1.6366,
      "step": 51
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.716003343634799e-05,
      "loss": 1.8304,
      "step": 52
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.724275869600789e-05,
      "loss": 1.6988,
      "step": 53
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7323937598229687e-05,
      "loss": 1.5348,
      "step": 54
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7403626894942437e-05,
      "loss": 1.7351,
      "step": 55
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7481880270062003e-05,
      "loss": 1.8166,
      "step": 56
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7558748556724913e-05,
      "loss": 1.9427,
      "step": 57
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.763427993562937e-05,
      "loss": 1.5295,
      "step": 58
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7708520116421443e-05,
      "loss": 1.69,
      "step": 59
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7781512503836432e-05,
      "loss": 1.862,
      "step": 60
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7853298350107667e-05,
      "loss": 1.8459,
      "step": 61
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7923916894982537e-05,
      "loss": 1.8489,
      "step": 62
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.7993405494535815e-05,
      "loss": 1.7041,
      "step": 63
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.806179973983887e-05,
      "loss": 1.8849,
      "step": 64
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.8129133566428553e-05,
      "loss": 1.5136,
      "step": 65
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.8195439355418686e-05,
      "loss": 1.7551,
      "step": 66
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.8260748027008263e-05,
      "loss": 1.673,
      "step": 67
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.8325089127062364e-05,
      "loss": 1.518,
      "step": 68
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8388490907372553e-05,
      "loss": 1.6828,
      "step": 69
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8450980400142568e-05,
      "loss": 1.5673,
      "step": 70
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8512583487190752e-05,
      "loss": 1.6589,
      "step": 71
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.857332496431268e-05,
      "loss": 1.8374,
      "step": 72
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8633228601204555e-05,
      "loss": 1.9306,
      "step": 73
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8692317197309763e-05,
      "loss": 1.8386,
      "step": 74
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8750612633916997e-05,
      "loss": 1.6398,
      "step": 75
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8808135922807914e-05,
      "loss": 1.7678,
      "step": 76
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8864907251724818e-05,
      "loss": 1.8326,
      "step": 77
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8920946026904802e-05,
      "loss": 1.8868,
      "step": 78
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.8976270912904414e-05,
      "loss": 1.5988,
      "step": 79
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9030899869919434e-05,
      "loss": 2.0657,
      "step": 80
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9084850188786497e-05,
      "loss": 1.7575,
      "step": 81
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9138138523837165e-05,
      "loss": 1.7669,
      "step": 82
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.919078092376074e-05,
      "loss": 1.8031,
      "step": 83
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9242792860618813e-05,
      "loss": 1.6363,
      "step": 84
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.929418925714293e-05,
      "loss": 1.5035,
      "step": 85
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9344984512435673e-05,
      "loss": 1.7458,
      "step": 86
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9395192526186183e-05,
      "loss": 1.5708,
      "step": 87
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9444826721501687e-05,
      "loss": 1.5143,
      "step": 88
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9493900066449125e-05,
      "loss": 1.7994,
      "step": 89
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9542425094393246e-05,
      "loss": 1.7253,
      "step": 90
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9590413923210934e-05,
      "loss": 1.796,
      "step": 91
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9637878273455555e-05,
      "loss": 1.8051,
      "step": 92
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.968482948553935e-05,
      "loss": 1.7542,
      "step": 93
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9731278535996986e-05,
      "loss": 1.5261,
      "step": 94
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9777236052888476e-05,
      "loss": 1.5642,
      "step": 95
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9822712330395683e-05,
      "loss": 1.719,
      "step": 96
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.986771734266245e-05,
      "loss": 1.6303,
      "step": 97
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.991226075692495e-05,
      "loss": 1.7,
      "step": 98
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9956351945975496e-05,
      "loss": 1.7116,
      "step": 99
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 1.7684,
      "step": 100
    },
    {
      "epoch": 0.02,
      "learning_rate": 2e-05,
      "loss": 1.7258,
      "step": 101
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.999851906701222e-05,
      "loss": 1.5624,
      "step": 102
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9997038134024437e-05,
      "loss": 1.6632,
      "step": 103
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9995557201036654e-05,
      "loss": 1.5455,
      "step": 104
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9994076268048872e-05,
      "loss": 1.5402,
      "step": 105
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.999259533506109e-05,
      "loss": 1.4711,
      "step": 106
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9991114402073307e-05,
      "loss": 1.8151,
      "step": 107
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9989633469085524e-05,
      "loss": 1.84,
      "step": 108
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9988152536097742e-05,
      "loss": 1.483,
      "step": 109
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9986671603109963e-05,
      "loss": 1.6337,
      "step": 110
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.998519067012218e-05,
      "loss": 1.9119,
      "step": 111
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9983709737134394e-05,
      "loss": 1.8984,
      "step": 112
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.9982228804146615e-05,
      "loss": 1.9073,
      "step": 113
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9980747871158833e-05,
      "loss": 1.5793,
      "step": 114
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.997926693817105e-05,
      "loss": 1.6775,
      "step": 115
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9977786005183268e-05,
      "loss": 1.9119,
      "step": 116
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9976305072195485e-05,
      "loss": 1.6897,
      "step": 117
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9974824139207703e-05,
      "loss": 1.7414,
      "step": 118
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.997334320621992e-05,
      "loss": 1.9912,
      "step": 119
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9971862273232138e-05,
      "loss": 1.5469,
      "step": 120
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9970381340244355e-05,
      "loss": 1.6811,
      "step": 121
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9968900407256573e-05,
      "loss": 1.6592,
      "step": 122
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.996741947426879e-05,
      "loss": 1.6979,
      "step": 123
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9965938541281008e-05,
      "loss": 1.5161,
      "step": 124
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9964457608293225e-05,
      "loss": 1.7055,
      "step": 125
    },
    {
      "epoch": 0.03,
      "eval_loss": NaN,
      "eval_runtime": 4.4628,
      "eval_samples_per_second": 345.972,
      "eval_steps_per_second": 43.246,
      "step": 125
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9962976675305443e-05,
      "loss": 1.576,
      "step": 126
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9961495742317664e-05,
      "loss": 1.6886,
      "step": 127
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9960014809329878e-05,
      "loss": 1.6073,
      "step": 128
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9958533876342095e-05,
      "loss": 1.9575,
      "step": 129
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9957052943354316e-05,
      "loss": 1.6877,
      "step": 130
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9955572010366534e-05,
      "loss": 1.9477,
      "step": 131
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.995409107737875e-05,
      "loss": 1.768,
      "step": 132
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9952610144390966e-05,
      "loss": 1.6431,
      "step": 133
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9951129211403186e-05,
      "loss": 2.0428,
      "step": 134
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9949648278415404e-05,
      "loss": 1.7046,
      "step": 135
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.994816734542762e-05,
      "loss": 1.7904,
      "step": 136
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.994668641243984e-05,
      "loss": 1.4176,
      "step": 137
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9945205479452057e-05,
      "loss": 1.5163,
      "step": 138
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9943724546464274e-05,
      "loss": 1.7338,
      "step": 139
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.994224361347649e-05,
      "loss": 1.5108,
      "step": 140
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.994076268048871e-05,
      "loss": 1.7425,
      "step": 141
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9939281747500927e-05,
      "loss": 1.6557,
      "step": 142
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9937800814513144e-05,
      "loss": 1.7534,
      "step": 143
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9936319881525365e-05,
      "loss": 1.5122,
      "step": 144
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.993483894853758e-05,
      "loss": 1.557,
      "step": 145
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9933358015549797e-05,
      "loss": 1.6656,
      "step": 146
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9931877082562018e-05,
      "loss": 1.6583,
      "step": 147
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9930396149574235e-05,
      "loss": 1.7867,
      "step": 148
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.992891521658645e-05,
      "loss": 1.4492,
      "step": 149
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9927434283598667e-05,
      "loss": 1.8965,
      "step": 150
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9925953350610888e-05,
      "loss": 1.5806,
      "step": 151
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9924472417623105e-05,
      "loss": 1.7388,
      "step": 152
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9922991484635323e-05,
      "loss": 1.8722,
      "step": 153
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.992151055164754e-05,
      "loss": 1.7304,
      "step": 154
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9920029618659758e-05,
      "loss": 1.5801,
      "step": 155
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9918548685671975e-05,
      "loss": 1.6197,
      "step": 156
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.9917067752684193e-05,
      "loss": 1.6764,
      "step": 157
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.991558681969641e-05,
      "loss": 1.8972,
      "step": 158
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9914105886708628e-05,
      "loss": 1.6693,
      "step": 159
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9912624953720845e-05,
      "loss": 1.6406,
      "step": 160
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9911144020733063e-05,
      "loss": 1.6621,
      "step": 161
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.990966308774528e-05,
      "loss": 1.9918,
      "step": 162
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9908182154757498e-05,
      "loss": 1.7839,
      "step": 163
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.990670122176972e-05,
      "loss": 1.8079,
      "step": 164
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9905220288781933e-05,
      "loss": 1.7007,
      "step": 165
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.990373935579415e-05,
      "loss": 1.6718,
      "step": 166
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9902258422806368e-05,
      "loss": 1.6458,
      "step": 167
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.990077748981859e-05,
      "loss": 1.5814,
      "step": 168
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9899296556830806e-05,
      "loss": 1.2563,
      "step": 169
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.989781562384302e-05,
      "loss": 1.6455,
      "step": 170
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.989633469085524e-05,
      "loss": 1.6263,
      "step": 171
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.989485375786746e-05,
      "loss": 1.7766,
      "step": 172
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9893372824879676e-05,
      "loss": 1.8603,
      "step": 173
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9891891891891894e-05,
      "loss": 1.9278,
      "step": 174
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.989041095890411e-05,
      "loss": 1.6557,
      "step": 175
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.988893002591633e-05,
      "loss": 1.7913,
      "step": 176
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9887449092928546e-05,
      "loss": 1.7318,
      "step": 177
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9885968159940764e-05,
      "loss": 1.5485,
      "step": 178
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.988448722695298e-05,
      "loss": 1.7427,
      "step": 179
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.98830062939652e-05,
      "loss": 1.4835,
      "step": 180
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.988152536097742e-05,
      "loss": 1.378,
      "step": 181
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9880044427989634e-05,
      "loss": 1.5273,
      "step": 182
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.987856349500185e-05,
      "loss": 1.378,
      "step": 183
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.987708256201407e-05,
      "loss": 1.4276,
      "step": 184
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.987560162902629e-05,
      "loss": 1.4386,
      "step": 185
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9874120696038504e-05,
      "loss": 1.5742,
      "step": 186
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.987263976305072e-05,
      "loss": 1.4334,
      "step": 187
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9871158830062943e-05,
      "loss": 1.7945,
      "step": 188
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.986967789707516e-05,
      "loss": 1.7118,
      "step": 189
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9868196964087378e-05,
      "loss": 1.7052,
      "step": 190
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9866716031099595e-05,
      "loss": 1.3457,
      "step": 191
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9865235098111813e-05,
      "loss": 1.619,
      "step": 192
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.986375416512403e-05,
      "loss": 1.7784,
      "step": 193
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9862273232136248e-05,
      "loss": 1.6964,
      "step": 194
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9860792299148465e-05,
      "loss": 1.5978,
      "step": 195
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9859311366160683e-05,
      "loss": 1.7352,
      "step": 196
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.98578304331729e-05,
      "loss": 1.4715,
      "step": 197
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9856349500185118e-05,
      "loss": 1.6171,
      "step": 198
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9854868567197335e-05,
      "loss": 1.572,
      "step": 199
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9853387634209553e-05,
      "loss": 1.9237,
      "step": 200
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.985190670122177e-05,
      "loss": 1.4677,
      "step": 201
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.985042576823399e-05,
      "loss": 1.4415,
      "step": 202
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9848944835246205e-05,
      "loss": 1.8609,
      "step": 203
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9847463902258423e-05,
      "loss": 1.6047,
      "step": 204
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9845982969270644e-05,
      "loss": 1.381,
      "step": 205
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.984450203628286e-05,
      "loss": 1.7174,
      "step": 206
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9843021103295075e-05,
      "loss": 1.753,
      "step": 207
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9841540170307296e-05,
      "loss": 1.5021,
      "step": 208
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9840059237319514e-05,
      "loss": 1.8122,
      "step": 209
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.983857830433173e-05,
      "loss": 1.4034,
      "step": 210
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.983709737134395e-05,
      "loss": 1.7725,
      "step": 211
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9835616438356166e-05,
      "loss": 1.8197,
      "step": 212
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9834135505368384e-05,
      "loss": 1.7396,
      "step": 213
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.98326545723806e-05,
      "loss": 1.6024,
      "step": 214
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.983117363939282e-05,
      "loss": 1.7314,
      "step": 215
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9829692706405036e-05,
      "loss": 1.5144,
      "step": 216
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9828211773417254e-05,
      "loss": 1.4619,
      "step": 217
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.982673084042947e-05,
      "loss": 1.9694,
      "step": 218
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.982524990744169e-05,
      "loss": 1.7189,
      "step": 219
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9823768974453906e-05,
      "loss": 1.5919,
      "step": 220
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9822288041466124e-05,
      "loss": 1.7036,
      "step": 221
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9820807108478345e-05,
      "loss": 1.7306,
      "step": 222
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9819326175490562e-05,
      "loss": 1.615,
      "step": 223
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9817845242502777e-05,
      "loss": 1.5968,
      "step": 224
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9816364309514997e-05,
      "loss": 1.6165,
      "step": 225
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9814883376527215e-05,
      "loss": 1.5267,
      "step": 226
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9813402443539432e-05,
      "loss": 1.7195,
      "step": 227
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9811921510551647e-05,
      "loss": 1.7924,
      "step": 228
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9810440577563868e-05,
      "loss": 1.5021,
      "step": 229
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9808959644576085e-05,
      "loss": 1.4931,
      "step": 230
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9807478711588303e-05,
      "loss": 1.7936,
      "step": 231
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.980599777860052e-05,
      "loss": 1.7176,
      "step": 232
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9804516845612738e-05,
      "loss": 1.6091,
      "step": 233
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9803035912624955e-05,
      "loss": 1.7259,
      "step": 234
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9801554979637173e-05,
      "loss": 1.6223,
      "step": 235
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.980007404664939e-05,
      "loss": 1.6864,
      "step": 236
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9798593113661608e-05,
      "loss": 1.6398,
      "step": 237
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9797112180673825e-05,
      "loss": 1.674,
      "step": 238
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9795631247686046e-05,
      "loss": 1.6472,
      "step": 239
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.979415031469826e-05,
      "loss": 1.5637,
      "step": 240
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9792669381710478e-05,
      "loss": 1.4302,
      "step": 241
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.97911884487227e-05,
      "loss": 1.5416,
      "step": 242
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9789707515734916e-05,
      "loss": 1.6245,
      "step": 243
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9788226582747134e-05,
      "loss": 1.5905,
      "step": 244
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9786745649759348e-05,
      "loss": 1.643,
      "step": 245
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.978526471677157e-05,
      "loss": 1.7263,
      "step": 246
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9783783783783786e-05,
      "loss": 1.6706,
      "step": 247
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9782302850796004e-05,
      "loss": 1.4921,
      "step": 248
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.978082191780822e-05,
      "loss": 1.7309,
      "step": 249
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.977934098482044e-05,
      "loss": 1.5623,
      "step": 250
    },
    {
      "epoch": 0.06,
      "eval_loss": NaN,
      "eval_runtime": 4.3752,
      "eval_samples_per_second": 352.901,
      "eval_steps_per_second": 44.113,
      "step": 250
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9777860051832656e-05,
      "loss": 1.4511,
      "step": 251
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9776379118844874e-05,
      "loss": 1.6214,
      "step": 252
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.977489818585709e-05,
      "loss": 1.3629,
      "step": 253
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.977341725286931e-05,
      "loss": 1.6919,
      "step": 254
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9771936319881526e-05,
      "loss": 1.7277,
      "step": 255
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9770455386893744e-05,
      "loss": 1.7145,
      "step": 256
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.976897445390596e-05,
      "loss": 1.6746,
      "step": 257
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.976749352091818e-05,
      "loss": 1.7496,
      "step": 258
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.97660125879304e-05,
      "loss": 1.623,
      "step": 259
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9764531654942617e-05,
      "loss": 1.78,
      "step": 260
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.976305072195483e-05,
      "loss": 1.6954,
      "step": 261
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.976156978896705e-05,
      "loss": 1.4804,
      "step": 262
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.976008885597927e-05,
      "loss": 1.8241,
      "step": 263
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9758607922991487e-05,
      "loss": 1.8181,
      "step": 264
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9757126990003705e-05,
      "loss": 1.6935,
      "step": 265
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9755646057015922e-05,
      "loss": 1.4718,
      "step": 266
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.975416512402814e-05,
      "loss": 1.4958,
      "step": 267
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9752684191040357e-05,
      "loss": 1.6213,
      "step": 268
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9751203258052575e-05,
      "loss": 1.7056,
      "step": 269
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9749722325064792e-05,
      "loss": 1.4955,
      "step": 270
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.974824139207701e-05,
      "loss": 1.553,
      "step": 271
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9746760459089228e-05,
      "loss": 1.6161,
      "step": 272
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9745279526101445e-05,
      "loss": 1.6912,
      "step": 273
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9743798593113663e-05,
      "loss": 1.5806,
      "step": 274
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.974231766012588e-05,
      "loss": 1.4892,
      "step": 275
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.97408367271381e-05,
      "loss": 1.5634,
      "step": 276
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9739355794150315e-05,
      "loss": 1.7621,
      "step": 277
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9737874861162533e-05,
      "loss": 1.7931,
      "step": 278
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.973639392817475e-05,
      "loss": 1.7043,
      "step": 279
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.973491299518697e-05,
      "loss": 1.6338,
      "step": 280
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.973343206219919e-05,
      "loss": 1.5115,
      "step": 281
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9731951129211403e-05,
      "loss": 1.7423,
      "step": 282
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9730470196223624e-05,
      "loss": 1.5749,
      "step": 283
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.972898926323584e-05,
      "loss": 1.6571,
      "step": 284
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.972750833024806e-05,
      "loss": 1.7007,
      "step": 285
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9726027397260276e-05,
      "loss": 1.4259,
      "step": 286
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9724546464272494e-05,
      "loss": 1.4144,
      "step": 287
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.972306553128471e-05,
      "loss": 1.6361,
      "step": 288
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.972158459829693e-05,
      "loss": 1.49,
      "step": 289
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9720103665309146e-05,
      "loss": 1.7114,
      "step": 290
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9718622732321364e-05,
      "loss": 1.7201,
      "step": 291
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.971714179933358e-05,
      "loss": 1.4985,
      "step": 292
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.97156608663458e-05,
      "loss": 1.7148,
      "step": 293
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.9714179933358016e-05,
      "loss": 1.6008,
      "step": 294
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9712699000370234e-05,
      "loss": 1.8057,
      "step": 295
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.971121806738245e-05,
      "loss": 1.5583,
      "step": 296
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9709737134394672e-05,
      "loss": 1.5456,
      "step": 297
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9708256201406886e-05,
      "loss": 1.6166,
      "step": 298
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9706775268419104e-05,
      "loss": 1.5364,
      "step": 299
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9705294335431325e-05,
      "loss": 1.6108,
      "step": 300
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9703813402443542e-05,
      "loss": 1.8692,
      "step": 301
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.970233246945576e-05,
      "loss": 1.7876,
      "step": 302
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9700851536467977e-05,
      "loss": 1.9047,
      "step": 303
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9699370603480195e-05,
      "loss": 1.7675,
      "step": 304
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9697889670492412e-05,
      "loss": 1.459,
      "step": 305
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.969640873750463e-05,
      "loss": 1.8272,
      "step": 306
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9694927804516847e-05,
      "loss": 1.6377,
      "step": 307
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9693446871529065e-05,
      "loss": 1.707,
      "step": 308
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9691965938541282e-05,
      "loss": 1.7615,
      "step": 309
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.96904850055535e-05,
      "loss": 1.6804,
      "step": 310
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9689004072565717e-05,
      "loss": 1.4299,
      "step": 311
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9687523139577935e-05,
      "loss": 1.2435,
      "step": 312
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9686042206590152e-05,
      "loss": 1.5995,
      "step": 313
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.968456127360237e-05,
      "loss": 1.6119,
      "step": 314
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9683080340614588e-05,
      "loss": 1.376,
      "step": 315
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9681599407626805e-05,
      "loss": 1.75,
      "step": 316
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9680118474639026e-05,
      "loss": 1.4592,
      "step": 317
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9678637541651243e-05,
      "loss": 1.3593,
      "step": 318
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9677156608663458e-05,
      "loss": 1.4297,
      "step": 319
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.967567567567568e-05,
      "loss": 1.6454,
      "step": 320
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9674194742687896e-05,
      "loss": 1.4377,
      "step": 321
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9672713809700114e-05,
      "loss": 1.5588,
      "step": 322
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.967123287671233e-05,
      "loss": 1.6108,
      "step": 323
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.966975194372455e-05,
      "loss": 1.6074,
      "step": 324
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9668271010736766e-05,
      "loss": 1.5041,
      "step": 325
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9666790077748984e-05,
      "loss": 1.5244,
      "step": 326
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.96653091447612e-05,
      "loss": 1.5515,
      "step": 327
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.966382821177342e-05,
      "loss": 1.5871,
      "step": 328
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9662347278785636e-05,
      "loss": 1.6337,
      "step": 329
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9660866345797854e-05,
      "loss": 1.4918,
      "step": 330
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.965938541281007e-05,
      "loss": 1.6649,
      "step": 331
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.965790447982229e-05,
      "loss": 1.5301,
      "step": 332
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9656423546834506e-05,
      "loss": 1.676,
      "step": 333
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9654942613846727e-05,
      "loss": 1.5394,
      "step": 334
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.965346168085894e-05,
      "loss": 1.6129,
      "step": 335
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.965198074787116e-05,
      "loss": 1.7893,
      "step": 336
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.965049981488338e-05,
      "loss": 1.4664,
      "step": 337
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9649018881895597e-05,
      "loss": 1.8448,
      "step": 338
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9647537948907815e-05,
      "loss": 1.9312,
      "step": 339
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.964605701592003e-05,
      "loss": 1.8983,
      "step": 340
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.964457608293225e-05,
      "loss": 1.4121,
      "step": 341
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9643095149944467e-05,
      "loss": 1.5184,
      "step": 342
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9641614216956685e-05,
      "loss": 1.7342,
      "step": 343
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9640133283968902e-05,
      "loss": 1.7112,
      "step": 344
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.963865235098112e-05,
      "loss": 1.48,
      "step": 345
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9637171417993337e-05,
      "loss": 1.6067,
      "step": 346
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9635690485005555e-05,
      "loss": 1.3944,
      "step": 347
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9634209552017772e-05,
      "loss": 1.4204,
      "step": 348
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.963272861902999e-05,
      "loss": 1.4149,
      "step": 349
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9631247686042207e-05,
      "loss": 1.6185,
      "step": 350
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9629766753054428e-05,
      "loss": 1.489,
      "step": 351
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9628285820066642e-05,
      "loss": 1.551,
      "step": 352
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.962680488707886e-05,
      "loss": 1.7149,
      "step": 353
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.962532395409108e-05,
      "loss": 1.7495,
      "step": 354
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.96238430211033e-05,
      "loss": 1.4699,
      "step": 355
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9622362088115512e-05,
      "loss": 1.7985,
      "step": 356
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.962088115512773e-05,
      "loss": 1.5384,
      "step": 357
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.961940022213995e-05,
      "loss": 1.6261,
      "step": 358
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.961791928915217e-05,
      "loss": 1.4456,
      "step": 359
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9616438356164386e-05,
      "loss": 1.5885,
      "step": 360
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9614957423176603e-05,
      "loss": 1.809,
      "step": 361
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.961347649018882e-05,
      "loss": 1.554,
      "step": 362
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.961199555720104e-05,
      "loss": 1.6679,
      "step": 363
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9610514624213256e-05,
      "loss": 1.5508,
      "step": 364
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9609033691225474e-05,
      "loss": 1.5844,
      "step": 365
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.960755275823769e-05,
      "loss": 1.6602,
      "step": 366
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.960607182524991e-05,
      "loss": 1.5476,
      "step": 367
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9604590892262126e-05,
      "loss": 1.6492,
      "step": 368
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9603109959274344e-05,
      "loss": 1.6276,
      "step": 369
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.960162902628656e-05,
      "loss": 1.5188,
      "step": 370
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9600148093298782e-05,
      "loss": 1.6088,
      "step": 371
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9598667160311e-05,
      "loss": 1.6344,
      "step": 372
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9597186227323214e-05,
      "loss": 1.5511,
      "step": 373
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.959570529433543e-05,
      "loss": 1.5812,
      "step": 374
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9594224361347652e-05,
      "loss": 1.5045,
      "step": 375
    },
    {
      "epoch": 0.08,
      "eval_loss": NaN,
      "eval_runtime": 4.3635,
      "eval_samples_per_second": 353.848,
      "eval_steps_per_second": 44.231,
      "step": 375
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.959274342835987e-05,
      "loss": 1.5198,
      "step": 376
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9591262495372084e-05,
      "loss": 1.5839,
      "step": 377
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9589781562384305e-05,
      "loss": 1.5428,
      "step": 378
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9588300629396522e-05,
      "loss": 1.5756,
      "step": 379
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.958681969640874e-05,
      "loss": 1.5859,
      "step": 380
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9585338763420957e-05,
      "loss": 1.4806,
      "step": 381
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9583857830433175e-05,
      "loss": 1.5064,
      "step": 382
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9582376897445392e-05,
      "loss": 1.8216,
      "step": 383
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.958089596445761e-05,
      "loss": 1.4468,
      "step": 384
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9579415031469827e-05,
      "loss": 1.7416,
      "step": 385
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9577934098482045e-05,
      "loss": 1.5334,
      "step": 386
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9576453165494262e-05,
      "loss": 1.0845,
      "step": 387
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9574972232506483e-05,
      "loss": 1.7763,
      "step": 388
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9573491299518697e-05,
      "loss": 1.4291,
      "step": 389
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9572010366530915e-05,
      "loss": 1.5607,
      "step": 390
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9570529433543132e-05,
      "loss": 1.6283,
      "step": 391
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9569048500555353e-05,
      "loss": 1.8019,
      "step": 392
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.956756756756757e-05,
      "loss": 1.5146,
      "step": 393
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9566086634579785e-05,
      "loss": 1.4971,
      "step": 394
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9564605701592006e-05,
      "loss": 1.5757,
      "step": 395
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9563124768604223e-05,
      "loss": 1.5229,
      "step": 396
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.956164383561644e-05,
      "loss": 1.515,
      "step": 397
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9560162902628655e-05,
      "loss": 1.7011,
      "step": 398
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9558681969640876e-05,
      "loss": 1.5874,
      "step": 399
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9557201036653093e-05,
      "loss": 1.5002,
      "step": 400
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.955572010366531e-05,
      "loss": 1.5118,
      "step": 401
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.955423917067753e-05,
      "loss": 1.8064,
      "step": 402
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9552758237689746e-05,
      "loss": 1.649,
      "step": 403
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9551277304701963e-05,
      "loss": 1.7451,
      "step": 404
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.954979637171418e-05,
      "loss": 1.6613,
      "step": 405
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.95483154387264e-05,
      "loss": 1.6721,
      "step": 406
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9546834505738616e-05,
      "loss": 1.487,
      "step": 407
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9545353572750834e-05,
      "loss": 1.5488,
      "step": 408
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9543872639763054e-05,
      "loss": 1.7579,
      "step": 409
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.954239170677527e-05,
      "loss": 1.6708,
      "step": 410
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9540910773787486e-05,
      "loss": 1.596,
      "step": 411
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9539429840799707e-05,
      "loss": 1.4587,
      "step": 412
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9537948907811925e-05,
      "loss": 1.6616,
      "step": 413
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9536467974824142e-05,
      "loss": 1.4619,
      "step": 414
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9534987041836356e-05,
      "loss": 1.6545,
      "step": 415
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9533506108848577e-05,
      "loss": 1.5677,
      "step": 416
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9532025175860795e-05,
      "loss": 1.4194,
      "step": 417
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9530544242873012e-05,
      "loss": 1.3929,
      "step": 418
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.952906330988523e-05,
      "loss": 1.8199,
      "step": 419
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9527582376897447e-05,
      "loss": 1.745,
      "step": 420
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9526101443909665e-05,
      "loss": 1.8072,
      "step": 421
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9524620510921882e-05,
      "loss": 1.4858,
      "step": 422
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.95231395779341e-05,
      "loss": 1.4978,
      "step": 423
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9521658644946317e-05,
      "loss": 1.6345,
      "step": 424
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9520177711958535e-05,
      "loss": 1.627,
      "step": 425
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9518696778970752e-05,
      "loss": 1.635,
      "step": 426
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.951721584598297e-05,
      "loss": 1.7005,
      "step": 427
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9515734912995187e-05,
      "loss": 1.5536,
      "step": 428
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9514253980007408e-05,
      "loss": 1.7383,
      "step": 429
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.9512773047019626e-05,
      "loss": 1.6188,
      "step": 430
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.951129211403184e-05,
      "loss": 1.6592,
      "step": 431
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9509811181044057e-05,
      "loss": 1.4564,
      "step": 432
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9508330248056278e-05,
      "loss": 1.5277,
      "step": 433
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9506849315068496e-05,
      "loss": 1.3895,
      "step": 434
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9505368382080713e-05,
      "loss": 1.5637,
      "step": 435
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.950388744909293e-05,
      "loss": 1.7683,
      "step": 436
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9502406516105148e-05,
      "loss": 1.5307,
      "step": 437
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9500925583117366e-05,
      "loss": 1.7226,
      "step": 438
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9499444650129583e-05,
      "loss": 2.0257,
      "step": 439
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.94979637171418e-05,
      "loss": 1.9701,
      "step": 440
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.949648278415402e-05,
      "loss": 1.7994,
      "step": 441
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9495001851166236e-05,
      "loss": 1.3895,
      "step": 442
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9493520918178453e-05,
      "loss": 1.5413,
      "step": 443
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.949203998519067e-05,
      "loss": 1.6755,
      "step": 444
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.949055905220289e-05,
      "loss": 1.5507,
      "step": 445
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.948907811921511e-05,
      "loss": 1.3068,
      "step": 446
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9487597186227323e-05,
      "loss": 1.5408,
      "step": 447
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.948611625323954e-05,
      "loss": 1.3767,
      "step": 448
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9484635320251762e-05,
      "loss": 1.7871,
      "step": 449
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.948315438726398e-05,
      "loss": 1.4614,
      "step": 450
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9481673454276197e-05,
      "loss": 1.7188,
      "step": 451
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.948019252128841e-05,
      "loss": 1.536,
      "step": 452
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9478711588300632e-05,
      "loss": 1.5591,
      "step": 453
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.947723065531285e-05,
      "loss": 1.4279,
      "step": 454
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9475749722325067e-05,
      "loss": 1.6069,
      "step": 455
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9474268789337285e-05,
      "loss": 1.6352,
      "step": 456
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9472787856349502e-05,
      "loss": 1.4507,
      "step": 457
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.947130692336172e-05,
      "loss": 1.5495,
      "step": 458
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9469825990373937e-05,
      "loss": 1.6701,
      "step": 459
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9468345057386155e-05,
      "loss": 1.3885,
      "step": 460
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9466864124398372e-05,
      "loss": 1.4798,
      "step": 461
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.946538319141059e-05,
      "loss": 1.6111,
      "step": 462
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9463902258422807e-05,
      "loss": 1.4785,
      "step": 463
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9462421325435025e-05,
      "loss": 1.4336,
      "step": 464
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9460940392447242e-05,
      "loss": 1.4635,
      "step": 465
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9459459459459463e-05,
      "loss": 1.5502,
      "step": 466
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.945797852647168e-05,
      "loss": 1.754,
      "step": 467
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9456497593483895e-05,
      "loss": 1.6609,
      "step": 468
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9455016660496112e-05,
      "loss": 1.6946,
      "step": 469
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9453535727508333e-05,
      "loss": 1.5559,
      "step": 470
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.945205479452055e-05,
      "loss": 1.442,
      "step": 471
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9450573861532768e-05,
      "loss": 1.5433,
      "step": 472
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9449092928544986e-05,
      "loss": 1.4958,
      "step": 473
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9447611995557203e-05,
      "loss": 1.6038,
      "step": 474
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.944613106256942e-05,
      "loss": 1.6523,
      "step": 475
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9444650129581638e-05,
      "loss": 1.5586,
      "step": 476
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9443169196593856e-05,
      "loss": 1.3989,
      "step": 477
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9441688263606073e-05,
      "loss": 1.3812,
      "step": 478
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.944020733061829e-05,
      "loss": 1.5716,
      "step": 479
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9438726397630508e-05,
      "loss": 1.553,
      "step": 480
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9437245464642726e-05,
      "loss": 1.4619,
      "step": 481
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9435764531654943e-05,
      "loss": 1.5058,
      "step": 482
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9434283598667164e-05,
      "loss": 1.8219,
      "step": 483
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.943280266567938e-05,
      "loss": 1.5152,
      "step": 484
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9431321732691596e-05,
      "loss": 1.4007,
      "step": 485
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9429840799703813e-05,
      "loss": 1.3942,
      "step": 486
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9428359866716034e-05,
      "loss": 1.6259,
      "step": 487
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9426878933728252e-05,
      "loss": 1.6042,
      "step": 488
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9425398000740466e-05,
      "loss": 1.4639,
      "step": 489
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9423917067752687e-05,
      "loss": 1.7573,
      "step": 490
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9422436134764904e-05,
      "loss": 1.7665,
      "step": 491
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9420955201777122e-05,
      "loss": 1.6637,
      "step": 492
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.941947426878934e-05,
      "loss": 1.5008,
      "step": 493
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9417993335801557e-05,
      "loss": 1.4926,
      "step": 494
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9416512402813774e-05,
      "loss": 1.472,
      "step": 495
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9415031469825992e-05,
      "loss": 1.6258,
      "step": 496
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.941355053683821e-05,
      "loss": 1.4844,
      "step": 497
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9412069603850427e-05,
      "loss": 1.3858,
      "step": 498
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9410588670862645e-05,
      "loss": 1.7216,
      "step": 499
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9409107737874865e-05,
      "loss": 1.8119,
      "step": 500
    },
    {
      "epoch": 0.11,
      "eval_loss": NaN,
      "eval_runtime": 4.3732,
      "eval_samples_per_second": 353.061,
      "eval_steps_per_second": 44.133,
      "step": 500
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.940762680488708e-05,
      "loss": 1.8826,
      "step": 501
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9406145871899297e-05,
      "loss": 1.3351,
      "step": 502
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9404664938911515e-05,
      "loss": 1.5995,
      "step": 503
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9403184005923735e-05,
      "loss": 1.4951,
      "step": 504
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.940170307293595e-05,
      "loss": 1.555,
      "step": 505
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9400222139948167e-05,
      "loss": 1.353,
      "step": 506
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9398741206960388e-05,
      "loss": 1.3169,
      "step": 507
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9397260273972606e-05,
      "loss": 1.4432,
      "step": 508
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9395779340984823e-05,
      "loss": 1.8469,
      "step": 509
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9394298407997037e-05,
      "loss": 1.6838,
      "step": 510
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9392817475009258e-05,
      "loss": 1.3505,
      "step": 511
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9391336542021476e-05,
      "loss": 1.4589,
      "step": 512
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9389855609033693e-05,
      "loss": 1.1256,
      "step": 513
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.938837467604591e-05,
      "loss": 1.6284,
      "step": 514
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9386893743058128e-05,
      "loss": 1.682,
      "step": 515
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9385412810070346e-05,
      "loss": 1.3105,
      "step": 516
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9383931877082563e-05,
      "loss": 1.6816,
      "step": 517
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.938245094409478e-05,
      "loss": 1.7434,
      "step": 518
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9380970011106998e-05,
      "loss": 1.6059,
      "step": 519
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9379489078119216e-05,
      "loss": 1.6146,
      "step": 520
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9378008145131437e-05,
      "loss": 1.8101,
      "step": 521
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.937652721214365e-05,
      "loss": 1.5285,
      "step": 522
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9375046279155868e-05,
      "loss": 1.5036,
      "step": 523
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.937356534616809e-05,
      "loss": 1.4642,
      "step": 524
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9372084413180307e-05,
      "loss": 1.5556,
      "step": 525
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.937060348019252e-05,
      "loss": 1.4223,
      "step": 526
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.936912254720474e-05,
      "loss": 1.5187,
      "step": 527
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.936764161421696e-05,
      "loss": 1.6771,
      "step": 528
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9366160681229177e-05,
      "loss": 1.5769,
      "step": 529
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9364679748241394e-05,
      "loss": 1.6663,
      "step": 530
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9363198815253612e-05,
      "loss": 1.6019,
      "step": 531
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.936171788226583e-05,
      "loss": 1.4003,
      "step": 532
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9360236949278047e-05,
      "loss": 1.5423,
      "step": 533
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9358756016290264e-05,
      "loss": 1.4825,
      "step": 534
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9357275083302482e-05,
      "loss": 1.2768,
      "step": 535
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.93557941503147e-05,
      "loss": 1.8682,
      "step": 536
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9354313217326917e-05,
      "loss": 1.5816,
      "step": 537
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9352832284339134e-05,
      "loss": 1.4334,
      "step": 538
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9351351351351352e-05,
      "loss": 1.6511,
      "step": 539
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.934987041836357e-05,
      "loss": 1.5453,
      "step": 540
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.934838948537579e-05,
      "loss": 1.4846,
      "step": 541
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9346908552388008e-05,
      "loss": 1.5639,
      "step": 542
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9345427619400222e-05,
      "loss": 1.4741,
      "step": 543
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.934394668641244e-05,
      "loss": 1.3409,
      "step": 544
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.934246575342466e-05,
      "loss": 1.4802,
      "step": 545
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9340984820436878e-05,
      "loss": 1.4598,
      "step": 546
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9339503887449092e-05,
      "loss": 1.5388,
      "step": 547
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9338022954461313e-05,
      "loss": 1.6268,
      "step": 548
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.933654202147353e-05,
      "loss": 1.434,
      "step": 549
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9335061088485748e-05,
      "loss": 1.472,
      "step": 550
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9333580155497966e-05,
      "loss": 1.5588,
      "step": 551
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9332099222510183e-05,
      "loss": 1.4413,
      "step": 552
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.93306182895224e-05,
      "loss": 1.4937,
      "step": 553
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9329137356534618e-05,
      "loss": 1.4249,
      "step": 554
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9327656423546836e-05,
      "loss": 1.5989,
      "step": 555
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9326175490559053e-05,
      "loss": 1.594,
      "step": 556
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.932469455757127e-05,
      "loss": 1.5449,
      "step": 557
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.932321362458349e-05,
      "loss": 1.4155,
      "step": 558
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9321732691595706e-05,
      "loss": 1.651,
      "step": 559
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9320251758607923e-05,
      "loss": 1.3055,
      "step": 560
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.931877082562014e-05,
      "loss": 1.529,
      "step": 561
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.931728989263236e-05,
      "loss": 1.4124,
      "step": 562
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.931580895964458e-05,
      "loss": 1.7339,
      "step": 563
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9314328026656793e-05,
      "loss": 1.6138,
      "step": 564
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9312847093669014e-05,
      "loss": 1.5231,
      "step": 565
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.9311366160681232e-05,
      "loss": 1.4383,
      "step": 566
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.930988522769345e-05,
      "loss": 1.6434,
      "step": 567
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9308404294705667e-05,
      "loss": 1.682,
      "step": 568
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9306923361717884e-05,
      "loss": 1.5121,
      "step": 569
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9305442428730102e-05,
      "loss": 1.6102,
      "step": 570
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.930396149574232e-05,
      "loss": 1.7233,
      "step": 571
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9302480562754537e-05,
      "loss": 1.5369,
      "step": 572
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9300999629766754e-05,
      "loss": 1.5325,
      "step": 573
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9299518696778972e-05,
      "loss": 1.6174,
      "step": 574
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.929803776379119e-05,
      "loss": 1.4448,
      "step": 575
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9296556830803407e-05,
      "loss": 1.4792,
      "step": 576
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9295075897815624e-05,
      "loss": 1.7899,
      "step": 577
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9293594964827845e-05,
      "loss": 1.5675,
      "step": 578
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9292114031840063e-05,
      "loss": 1.6407,
      "step": 579
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9290633098852277e-05,
      "loss": 1.425,
      "step": 580
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9289152165864494e-05,
      "loss": 1.5892,
      "step": 581
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9287671232876715e-05,
      "loss": 1.4374,
      "step": 582
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9286190299888933e-05,
      "loss": 1.6222,
      "step": 583
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.928470936690115e-05,
      "loss": 1.3517,
      "step": 584
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9283228433913368e-05,
      "loss": 1.4158,
      "step": 585
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9281747500925585e-05,
      "loss": 1.6741,
      "step": 586
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9280266567937803e-05,
      "loss": 1.77,
      "step": 587
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.927878563495002e-05,
      "loss": 1.4716,
      "step": 588
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9277304701962238e-05,
      "loss": 1.4776,
      "step": 589
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9275823768974455e-05,
      "loss": 1.4734,
      "step": 590
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9274342835986673e-05,
      "loss": 1.9138,
      "step": 591
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.927286190299889e-05,
      "loss": 1.8686,
      "step": 592
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9271380970011108e-05,
      "loss": 1.6448,
      "step": 593
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9269900037023326e-05,
      "loss": 1.7094,
      "step": 594
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9268419104035546e-05,
      "loss": 1.6585,
      "step": 595
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.926693817104776e-05,
      "loss": 1.5534,
      "step": 596
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9265457238059978e-05,
      "loss": 1.4579,
      "step": 597
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9263976305072196e-05,
      "loss": 1.504,
      "step": 598
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9262495372084417e-05,
      "loss": 1.534,
      "step": 599
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9261014439096634e-05,
      "loss": 1.606,
      "step": 600
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9259533506108848e-05,
      "loss": 1.4301,
      "step": 601
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.925805257312107e-05,
      "loss": 1.6601,
      "step": 602
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9256571640133287e-05,
      "loss": 1.5493,
      "step": 603
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9255090707145504e-05,
      "loss": 1.6509,
      "step": 604
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9253609774157718e-05,
      "loss": 1.6472,
      "step": 605
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.925212884116994e-05,
      "loss": 1.6657,
      "step": 606
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9250647908182157e-05,
      "loss": 1.6215,
      "step": 607
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9249166975194374e-05,
      "loss": 1.5097,
      "step": 608
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9247686042206592e-05,
      "loss": 1.6651,
      "step": 609
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.924620510921881e-05,
      "loss": 1.4343,
      "step": 610
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9244724176231027e-05,
      "loss": 1.6397,
      "step": 611
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9243243243243244e-05,
      "loss": 1.5168,
      "step": 612
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9241762310255462e-05,
      "loss": 1.7179,
      "step": 613
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.924028137726768e-05,
      "loss": 1.5243,
      "step": 614
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9238800444279897e-05,
      "loss": 1.6415,
      "step": 615
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9237319511292118e-05,
      "loss": 1.3047,
      "step": 616
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9235838578304332e-05,
      "loss": 1.5428,
      "step": 617
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.923435764531655e-05,
      "loss": 1.4271,
      "step": 618
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.923287671232877e-05,
      "loss": 1.5863,
      "step": 619
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9231395779340988e-05,
      "loss": 1.6041,
      "step": 620
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9229914846353205e-05,
      "loss": 1.4976,
      "step": 621
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.922843391336542e-05,
      "loss": 1.6478,
      "step": 622
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.922695298037764e-05,
      "loss": 1.6437,
      "step": 623
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9225472047389858e-05,
      "loss": 1.9261,
      "step": 624
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9223991114402075e-05,
      "loss": 1.4225,
      "step": 625
    },
    {
      "epoch": 0.14,
      "eval_loss": NaN,
      "eval_runtime": 4.3777,
      "eval_samples_per_second": 352.694,
      "eval_steps_per_second": 44.087,
      "step": 625
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9222510181414293e-05,
      "loss": 1.3998,
      "step": 626
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.922102924842651e-05,
      "loss": 1.432,
      "step": 627
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9219548315438728e-05,
      "loss": 1.4413,
      "step": 628
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9218067382450945e-05,
      "loss": 1.6801,
      "step": 629
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9216586449463163e-05,
      "loss": 1.5764,
      "step": 630
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.921510551647538e-05,
      "loss": 1.4587,
      "step": 631
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9213624583487598e-05,
      "loss": 1.4449,
      "step": 632
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9212143650499815e-05,
      "loss": 1.6043,
      "step": 633
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9210662717512033e-05,
      "loss": 1.435,
      "step": 634
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.920918178452425e-05,
      "loss": 1.6428,
      "step": 635
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.920770085153647e-05,
      "loss": 1.4563,
      "step": 636
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.920621991854869e-05,
      "loss": 1.7433,
      "step": 637
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9204738985560903e-05,
      "loss": 1.8292,
      "step": 638
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.920325805257312e-05,
      "loss": 1.6363,
      "step": 639
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.920177711958534e-05,
      "loss": 1.364,
      "step": 640
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.920029618659756e-05,
      "loss": 1.395,
      "step": 641
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9198815253609777e-05,
      "loss": 1.7652,
      "step": 642
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9197334320621994e-05,
      "loss": 1.5681,
      "step": 643
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.919585338763421e-05,
      "loss": 1.4614,
      "step": 644
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.919437245464643e-05,
      "loss": 1.4241,
      "step": 645
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9192891521658647e-05,
      "loss": 1.6332,
      "step": 646
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9191410588670864e-05,
      "loss": 1.5773,
      "step": 647
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.918992965568308e-05,
      "loss": 1.6034,
      "step": 648
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.91884487226953e-05,
      "loss": 1.6409,
      "step": 649
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9186967789707517e-05,
      "loss": 1.5852,
      "step": 650
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9185486856719734e-05,
      "loss": 1.4421,
      "step": 651
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9184005923731952e-05,
      "loss": 1.6738,
      "step": 652
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9182524990744173e-05,
      "loss": 1.6374,
      "step": 653
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9181044057756387e-05,
      "loss": 1.4138,
      "step": 654
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9179563124768604e-05,
      "loss": 1.359,
      "step": 655
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9178082191780822e-05,
      "loss": 1.4799,
      "step": 656
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9176601258793043e-05,
      "loss": 1.6182,
      "step": 657
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.917512032580526e-05,
      "loss": 1.5807,
      "step": 658
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9173639392817474e-05,
      "loss": 1.5793,
      "step": 659
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9172158459829695e-05,
      "loss": 1.5752,
      "step": 660
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9170677526841913e-05,
      "loss": 1.6364,
      "step": 661
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.916919659385413e-05,
      "loss": 1.9077,
      "step": 662
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9167715660866348e-05,
      "loss": 1.3736,
      "step": 663
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9166234727878565e-05,
      "loss": 1.4165,
      "step": 664
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9164753794890783e-05,
      "loss": 1.5267,
      "step": 665
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9163272861903e-05,
      "loss": 1.4777,
      "step": 666
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9161791928915218e-05,
      "loss": 1.5044,
      "step": 667
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9160310995927435e-05,
      "loss": 1.5,
      "step": 668
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9158830062939653e-05,
      "loss": 1.806,
      "step": 669
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9157349129951874e-05,
      "loss": 1.6494,
      "step": 670
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9155868196964088e-05,
      "loss": 1.6043,
      "step": 671
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9154387263976305e-05,
      "loss": 1.5244,
      "step": 672
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9152906330988523e-05,
      "loss": 1.43,
      "step": 673
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9151425398000744e-05,
      "loss": 1.5201,
      "step": 674
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9149944465012958e-05,
      "loss": 1.5105,
      "step": 675
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9148463532025175e-05,
      "loss": 1.3535,
      "step": 676
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9146982599037396e-05,
      "loss": 1.5725,
      "step": 677
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9145501666049614e-05,
      "loss": 1.5966,
      "step": 678
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.914402073306183e-05,
      "loss": 1.4531,
      "step": 679
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.914253980007405e-05,
      "loss": 1.5976,
      "step": 680
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9141058867086266e-05,
      "loss": 1.6501,
      "step": 681
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9139577934098484e-05,
      "loss": 1.4268,
      "step": 682
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.91380970011107e-05,
      "loss": 1.7491,
      "step": 683
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.913661606812292e-05,
      "loss": 1.7035,
      "step": 684
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9135135135135137e-05,
      "loss": 1.7333,
      "step": 685
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9133654202147354e-05,
      "loss": 1.4139,
      "step": 686
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.913217326915957e-05,
      "loss": 1.4513,
      "step": 687
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.913069233617179e-05,
      "loss": 1.6721,
      "step": 688
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9129211403184007e-05,
      "loss": 1.4909,
      "step": 689
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9127730470196224e-05,
      "loss": 1.4968,
      "step": 690
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9126249537208445e-05,
      "loss": 1.7072,
      "step": 691
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.912476860422066e-05,
      "loss": 1.5095,
      "step": 692
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9123287671232877e-05,
      "loss": 1.6126,
      "step": 693
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9121806738245098e-05,
      "loss": 1.5322,
      "step": 694
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9120325805257315e-05,
      "loss": 1.4002,
      "step": 695
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.911884487226953e-05,
      "loss": 1.2933,
      "step": 696
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.911736393928175e-05,
      "loss": 1.6266,
      "step": 697
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9115883006293968e-05,
      "loss": 1.3831,
      "step": 698
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9114402073306185e-05,
      "loss": 1.4583,
      "step": 699
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9112921140318403e-05,
      "loss": 1.502,
      "step": 700
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.911144020733062e-05,
      "loss": 1.6084,
      "step": 701
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.9109959274342838e-05,
      "loss": 1.6508,
      "step": 702
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9108478341355055e-05,
      "loss": 1.6573,
      "step": 703
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9106997408367273e-05,
      "loss": 1.4728,
      "step": 704
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.910551647537949e-05,
      "loss": 1.6002,
      "step": 705
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9104035542391708e-05,
      "loss": 1.5601,
      "step": 706
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.910255460940393e-05,
      "loss": 1.2803,
      "step": 707
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9101073676416143e-05,
      "loss": 1.3312,
      "step": 708
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.909959274342836e-05,
      "loss": 1.7154,
      "step": 709
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9098111810440578e-05,
      "loss": 1.6132,
      "step": 710
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.90966308774528e-05,
      "loss": 1.7576,
      "step": 711
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9095149944465016e-05,
      "loss": 1.65,
      "step": 712
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.909366901147723e-05,
      "loss": 1.6856,
      "step": 713
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.909218807848945e-05,
      "loss": 1.6601,
      "step": 714
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.909070714550167e-05,
      "loss": 1.6972,
      "step": 715
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9089226212513886e-05,
      "loss": 1.4189,
      "step": 716
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.90877452795261e-05,
      "loss": 1.5881,
      "step": 717
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.908626434653832e-05,
      "loss": 1.5668,
      "step": 718
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.908478341355054e-05,
      "loss": 1.3173,
      "step": 719
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9083302480562756e-05,
      "loss": 1.4417,
      "step": 720
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9081821547574974e-05,
      "loss": 1.4349,
      "step": 721
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.908034061458719e-05,
      "loss": 1.7384,
      "step": 722
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.907885968159941e-05,
      "loss": 1.6981,
      "step": 723
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9077378748611626e-05,
      "loss": 1.719,
      "step": 724
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9075897815623844e-05,
      "loss": 1.5818,
      "step": 725
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.907441688263606e-05,
      "loss": 1.619,
      "step": 726
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.907293594964828e-05,
      "loss": 1.513,
      "step": 727
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.90714550166605e-05,
      "loss": 1.5972,
      "step": 728
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9069974083672714e-05,
      "loss": 1.681,
      "step": 729
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.906849315068493e-05,
      "loss": 1.3818,
      "step": 730
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9067012217697152e-05,
      "loss": 1.5806,
      "step": 731
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.906553128470937e-05,
      "loss": 1.3965,
      "step": 732
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9064050351721584e-05,
      "loss": 1.8595,
      "step": 733
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.90625694187338e-05,
      "loss": 1.4471,
      "step": 734
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9061088485746023e-05,
      "loss": 1.5163,
      "step": 735
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.905960755275824e-05,
      "loss": 1.3795,
      "step": 736
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9058126619770458e-05,
      "loss": 1.4273,
      "step": 737
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9056645686782675e-05,
      "loss": 1.6424,
      "step": 738
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9055164753794893e-05,
      "loss": 1.495,
      "step": 739
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.905368382080711e-05,
      "loss": 1.786,
      "step": 740
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9052202887819328e-05,
      "loss": 1.5456,
      "step": 741
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9050721954831545e-05,
      "loss": 1.3615,
      "step": 742
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9049241021843763e-05,
      "loss": 1.6119,
      "step": 743
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.904776008885598e-05,
      "loss": 1.4522,
      "step": 744
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9046279155868198e-05,
      "loss": 1.471,
      "step": 745
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9044798222880415e-05,
      "loss": 1.5057,
      "step": 746
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9043317289892633e-05,
      "loss": 1.643,
      "step": 747
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9041836356904854e-05,
      "loss": 1.8603,
      "step": 748
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.904035542391707e-05,
      "loss": 1.516,
      "step": 749
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9038874490929285e-05,
      "loss": 1.4297,
      "step": 750
    },
    {
      "epoch": 0.17,
      "eval_loss": NaN,
      "eval_runtime": 4.369,
      "eval_samples_per_second": 353.398,
      "eval_steps_per_second": 44.175,
      "step": 750
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9037393557941503e-05,
      "loss": 1.5054,
      "step": 751
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9035912624953724e-05,
      "loss": 1.5455,
      "step": 752
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.903443169196594e-05,
      "loss": 1.5702,
      "step": 753
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9032950758978155e-05,
      "loss": 1.2744,
      "step": 754
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9031469825990376e-05,
      "loss": 1.7749,
      "step": 755
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9029988893002594e-05,
      "loss": 1.3792,
      "step": 756
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.902850796001481e-05,
      "loss": 1.6178,
      "step": 757
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.902702702702703e-05,
      "loss": 1.4959,
      "step": 758
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9025546094039246e-05,
      "loss": 1.4694,
      "step": 759
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9024065161051464e-05,
      "loss": 1.659,
      "step": 760
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.902258422806368e-05,
      "loss": 1.5362,
      "step": 761
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.90211032950759e-05,
      "loss": 1.6311,
      "step": 762
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9019622362088116e-05,
      "loss": 1.6192,
      "step": 763
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9018141429100334e-05,
      "loss": 1.7294,
      "step": 764
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9016660496112555e-05,
      "loss": 1.6867,
      "step": 765
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.901517956312477e-05,
      "loss": 1.67,
      "step": 766
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9013698630136986e-05,
      "loss": 1.429,
      "step": 767
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9012217697149204e-05,
      "loss": 1.8141,
      "step": 768
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9010736764161425e-05,
      "loss": 1.7129,
      "step": 769
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9009255831173642e-05,
      "loss": 1.7106,
      "step": 770
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9007774898185857e-05,
      "loss": 1.6934,
      "step": 771
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9006293965198077e-05,
      "loss": 1.5862,
      "step": 772
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9004813032210295e-05,
      "loss": 1.7154,
      "step": 773
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9003332099222512e-05,
      "loss": 1.4971,
      "step": 774
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.900185116623473e-05,
      "loss": 1.4911,
      "step": 775
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.9000370233246948e-05,
      "loss": 1.5834,
      "step": 776
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8998889300259165e-05,
      "loss": 1.5611,
      "step": 777
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8997408367271383e-05,
      "loss": 1.7946,
      "step": 778
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.89959274342836e-05,
      "loss": 1.4755,
      "step": 779
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8994446501295818e-05,
      "loss": 1.7822,
      "step": 780
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8992965568308035e-05,
      "loss": 1.5849,
      "step": 781
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8991484635320253e-05,
      "loss": 1.4563,
      "step": 782
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.899000370233247e-05,
      "loss": 1.6245,
      "step": 783
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8988522769344688e-05,
      "loss": 1.5832,
      "step": 784
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8987041836356905e-05,
      "loss": 1.5946,
      "step": 785
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8985560903369126e-05,
      "loss": 1.4484,
      "step": 786
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.898407997038134e-05,
      "loss": 1.6472,
      "step": 787
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8982599037393558e-05,
      "loss": 1.4404,
      "step": 788
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.898111810440578e-05,
      "loss": 1.4861,
      "step": 789
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8979637171417996e-05,
      "loss": 1.6095,
      "step": 790
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.8978156238430214e-05,
      "loss": 1.5967,
      "step": 791
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.897667530544243e-05,
      "loss": 1.6615,
      "step": 792
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.897519437245465e-05,
      "loss": 1.6054,
      "step": 793
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8973713439466866e-05,
      "loss": 1.2626,
      "step": 794
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8972232506479084e-05,
      "loss": 1.6999,
      "step": 795
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.89707515734913e-05,
      "loss": 1.7453,
      "step": 796
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.896927064050352e-05,
      "loss": 1.6812,
      "step": 797
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8967789707515736e-05,
      "loss": 1.4716,
      "step": 798
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8966308774527954e-05,
      "loss": 1.6565,
      "step": 799
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.896482784154017e-05,
      "loss": 1.4793,
      "step": 800
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.896334690855239e-05,
      "loss": 1.3877,
      "step": 801
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8961865975564606e-05,
      "loss": 1.6575,
      "step": 802
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8960385042576824e-05,
      "loss": 1.6643,
      "step": 803
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.895890410958904e-05,
      "loss": 1.5864,
      "step": 804
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.895742317660126e-05,
      "loss": 1.6098,
      "step": 805
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.895594224361348e-05,
      "loss": 1.8054,
      "step": 806
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8954461310625697e-05,
      "loss": 1.538,
      "step": 807
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.895298037763791e-05,
      "loss": 1.3489,
      "step": 808
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8951499444650132e-05,
      "loss": 1.2354,
      "step": 809
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.895001851166235e-05,
      "loss": 1.3854,
      "step": 810
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8948537578674567e-05,
      "loss": 1.5396,
      "step": 811
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8947056645686785e-05,
      "loss": 1.6373,
      "step": 812
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8945575712699002e-05,
      "loss": 1.488,
      "step": 813
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.894409477971122e-05,
      "loss": 1.5938,
      "step": 814
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8942613846723437e-05,
      "loss": 1.4931,
      "step": 815
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8941132913735655e-05,
      "loss": 1.5843,
      "step": 816
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8939651980747872e-05,
      "loss": 1.7632,
      "step": 817
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.893817104776009e-05,
      "loss": 1.6608,
      "step": 818
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8936690114772308e-05,
      "loss": 1.5599,
      "step": 819
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8935209181784525e-05,
      "loss": 1.4059,
      "step": 820
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8933728248796743e-05,
      "loss": 1.6741,
      "step": 821
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.893224731580896e-05,
      "loss": 1.3421,
      "step": 822
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.893076638282118e-05,
      "loss": 1.3534,
      "step": 823
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8929285449833395e-05,
      "loss": 1.4753,
      "step": 824
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8927804516845613e-05,
      "loss": 1.6063,
      "step": 825
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8926323583857834e-05,
      "loss": 1.4808,
      "step": 826
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.892484265087005e-05,
      "loss": 1.6779,
      "step": 827
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.892336171788227e-05,
      "loss": 1.5301,
      "step": 828
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8921880784894483e-05,
      "loss": 1.5156,
      "step": 829
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8920399851906704e-05,
      "loss": 1.775,
      "step": 830
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.891891891891892e-05,
      "loss": 1.7115,
      "step": 831
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.891743798593114e-05,
      "loss": 1.5426,
      "step": 832
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8915957052943356e-05,
      "loss": 1.5916,
      "step": 833
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8914476119955574e-05,
      "loss": 1.578,
      "step": 834
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.891299518696779e-05,
      "loss": 1.5533,
      "step": 835
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.891151425398001e-05,
      "loss": 1.4539,
      "step": 836
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8910033320992226e-05,
      "loss": 1.7327,
      "step": 837
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.8908552388004444e-05,
      "loss": 1.6615,
      "step": 838
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.890707145501666e-05,
      "loss": 1.5487,
      "step": 839
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8905590522028882e-05,
      "loss": 1.5385,
      "step": 840
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8904109589041096e-05,
      "loss": 1.3128,
      "step": 841
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8902628656053314e-05,
      "loss": 1.7076,
      "step": 842
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8901147723065535e-05,
      "loss": 1.5632,
      "step": 843
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8899666790077752e-05,
      "loss": 1.5378,
      "step": 844
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8898185857089966e-05,
      "loss": 1.6142,
      "step": 845
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8896704924102184e-05,
      "loss": 1.6508,
      "step": 846
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8895223991114405e-05,
      "loss": 1.5864,
      "step": 847
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8893743058126622e-05,
      "loss": 1.6934,
      "step": 848
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.889226212513884e-05,
      "loss": 1.4338,
      "step": 849
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8890781192151057e-05,
      "loss": 1.5457,
      "step": 850
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8889300259163275e-05,
      "loss": 1.5447,
      "step": 851
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8887819326175492e-05,
      "loss": 1.3386,
      "step": 852
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.888633839318771e-05,
      "loss": 1.3004,
      "step": 853
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8884857460199927e-05,
      "loss": 1.3689,
      "step": 854
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8883376527212145e-05,
      "loss": 1.5646,
      "step": 855
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8881895594224362e-05,
      "loss": 1.3699,
      "step": 856
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.888041466123658e-05,
      "loss": 1.5148,
      "step": 857
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8878933728248797e-05,
      "loss": 1.4885,
      "step": 858
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8877452795261015e-05,
      "loss": 1.6351,
      "step": 859
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8875971862273236e-05,
      "loss": 1.7357,
      "step": 860
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.887449092928545e-05,
      "loss": 1.6568,
      "step": 861
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8873009996297668e-05,
      "loss": 1.4142,
      "step": 862
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8871529063309885e-05,
      "loss": 1.7332,
      "step": 863
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8870048130322106e-05,
      "loss": 1.579,
      "step": 864
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8868567197334323e-05,
      "loss": 1.424,
      "step": 865
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8867086264346538e-05,
      "loss": 1.6454,
      "step": 866
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.886560533135876e-05,
      "loss": 1.5904,
      "step": 867
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8864124398370976e-05,
      "loss": 1.5106,
      "step": 868
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8862643465383194e-05,
      "loss": 1.6361,
      "step": 869
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.886116253239541e-05,
      "loss": 1.6104,
      "step": 870
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.885968159940763e-05,
      "loss": 1.3299,
      "step": 871
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8858200666419846e-05,
      "loss": 1.5125,
      "step": 872
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8856719733432064e-05,
      "loss": 1.5604,
      "step": 873
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.885523880044428e-05,
      "loss": 1.4046,
      "step": 874
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.88537578674565e-05,
      "loss": 1.5772,
      "step": 875
    },
    {
      "epoch": 0.19,
      "eval_loss": NaN,
      "eval_runtime": 4.3833,
      "eval_samples_per_second": 352.248,
      "eval_steps_per_second": 44.031,
      "step": 875
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8852276934468716e-05,
      "loss": 1.5348,
      "step": 876
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8850796001480937e-05,
      "loss": 1.3127,
      "step": 877
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.884931506849315e-05,
      "loss": 1.4865,
      "step": 878
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.884783413550537e-05,
      "loss": 1.6891,
      "step": 879
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8846353202517586e-05,
      "loss": 1.6867,
      "step": 880
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.8844872269529807e-05,
      "loss": 1.5948,
      "step": 881
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.884339133654202e-05,
      "loss": 1.4355,
      "step": 882
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.884191040355424e-05,
      "loss": 1.6134,
      "step": 883
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.884042947056646e-05,
      "loss": 1.3518,
      "step": 884
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8838948537578677e-05,
      "loss": 1.4305,
      "step": 885
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8837467604590895e-05,
      "loss": 1.5559,
      "step": 886
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8835986671603112e-05,
      "loss": 1.5585,
      "step": 887
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.883450573861533e-05,
      "loss": 1.6004,
      "step": 888
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8833024805627547e-05,
      "loss": 1.4324,
      "step": 889
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8831543872639765e-05,
      "loss": 1.3771,
      "step": 890
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8830062939651982e-05,
      "loss": 1.7197,
      "step": 891
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.88285820066642e-05,
      "loss": 1.4033,
      "step": 892
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8827101073676417e-05,
      "loss": 1.592,
      "step": 893
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8825620140688635e-05,
      "loss": 1.4341,
      "step": 894
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8824139207700852e-05,
      "loss": 1.6366,
      "step": 895
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.882265827471307e-05,
      "loss": 1.5738,
      "step": 896
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8821177341725287e-05,
      "loss": 1.6864,
      "step": 897
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8819696408737508e-05,
      "loss": 1.6697,
      "step": 898
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8818215475749722e-05,
      "loss": 1.3994,
      "step": 899
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.881673454276194e-05,
      "loss": 1.5355,
      "step": 900
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.881525360977416e-05,
      "loss": 1.241,
      "step": 901
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.881377267678638e-05,
      "loss": 1.3446,
      "step": 902
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8812291743798592e-05,
      "loss": 1.5124,
      "step": 903
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8810810810810813e-05,
      "loss": 1.4643,
      "step": 904
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.880932987782303e-05,
      "loss": 1.533,
      "step": 905
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.880784894483525e-05,
      "loss": 1.6927,
      "step": 906
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8806368011847466e-05,
      "loss": 1.7843,
      "step": 907
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8804887078859683e-05,
      "loss": 1.2216,
      "step": 908
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.88034061458719e-05,
      "loss": 1.3414,
      "step": 909
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.880192521288412e-05,
      "loss": 1.4828,
      "step": 910
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8800444279896336e-05,
      "loss": 1.6242,
      "step": 911
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8798963346908554e-05,
      "loss": 1.4215,
      "step": 912
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.879748241392077e-05,
      "loss": 1.4119,
      "step": 913
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.879600148093299e-05,
      "loss": 1.7393,
      "step": 914
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8794520547945206e-05,
      "loss": 1.5435,
      "step": 915
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8793039614957424e-05,
      "loss": 1.4963,
      "step": 916
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.879155868196964e-05,
      "loss": 1.3979,
      "step": 917
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8790077748981862e-05,
      "loss": 1.3079,
      "step": 918
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.878859681599408e-05,
      "loss": 1.4214,
      "step": 919
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8787115883006294e-05,
      "loss": 1.49,
      "step": 920
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8785634950018515e-05,
      "loss": 1.1918,
      "step": 921
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8784154017030732e-05,
      "loss": 1.5149,
      "step": 922
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.878267308404295e-05,
      "loss": 1.6047,
      "step": 923
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8781192151055164e-05,
      "loss": 1.6837,
      "step": 924
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8779711218067385e-05,
      "loss": 1.5812,
      "step": 925
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8778230285079602e-05,
      "loss": 1.4704,
      "step": 926
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.877674935209182e-05,
      "loss": 1.3692,
      "step": 927
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8775268419104037e-05,
      "loss": 1.5718,
      "step": 928
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8773787486116255e-05,
      "loss": 1.4812,
      "step": 929
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8772306553128472e-05,
      "loss": 1.5099,
      "step": 930
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.877082562014069e-05,
      "loss": 1.4317,
      "step": 931
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8769344687152907e-05,
      "loss": 1.5043,
      "step": 932
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8767863754165125e-05,
      "loss": 1.5552,
      "step": 933
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8766382821177342e-05,
      "loss": 1.5151,
      "step": 934
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8764901888189563e-05,
      "loss": 1.6228,
      "step": 935
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8763420955201777e-05,
      "loss": 1.4794,
      "step": 936
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8761940022213995e-05,
      "loss": 1.8094,
      "step": 937
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8760459089226216e-05,
      "loss": 1.5589,
      "step": 938
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8758978156238433e-05,
      "loss": 1.513,
      "step": 939
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.875749722325065e-05,
      "loss": 1.5445,
      "step": 940
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8756016290262865e-05,
      "loss": 1.5863,
      "step": 941
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8754535357275086e-05,
      "loss": 1.4394,
      "step": 942
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8753054424287303e-05,
      "loss": 1.2967,
      "step": 943
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.875157349129952e-05,
      "loss": 1.3676,
      "step": 944
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.875009255831174e-05,
      "loss": 1.3437,
      "step": 945
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8748611625323956e-05,
      "loss": 1.5894,
      "step": 946
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8747130692336173e-05,
      "loss": 1.421,
      "step": 947
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.874564975934839e-05,
      "loss": 1.5901,
      "step": 948
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.874416882636061e-05,
      "loss": 1.6067,
      "step": 949
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8742687893372826e-05,
      "loss": 1.5168,
      "step": 950
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8741206960385043e-05,
      "loss": 1.4606,
      "step": 951
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.873972602739726e-05,
      "loss": 1.3465,
      "step": 952
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.873824509440948e-05,
      "loss": 1.5604,
      "step": 953
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8736764161421696e-05,
      "loss": 1.4766,
      "step": 954
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8735283228433917e-05,
      "loss": 1.1961,
      "step": 955
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8733802295446134e-05,
      "loss": 1.52,
      "step": 956
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.873232136245835e-05,
      "loss": 1.4342,
      "step": 957
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8730840429470566e-05,
      "loss": 1.5385,
      "step": 958
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8729359496482787e-05,
      "loss": 1.6946,
      "step": 959
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8727878563495005e-05,
      "loss": 1.6476,
      "step": 960
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8726397630507222e-05,
      "loss": 1.545,
      "step": 961
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.872491669751944e-05,
      "loss": 1.376,
      "step": 962
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8723435764531657e-05,
      "loss": 1.4948,
      "step": 963
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8721954831543875e-05,
      "loss": 1.649,
      "step": 964
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8720473898556092e-05,
      "loss": 1.2994,
      "step": 965
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.871899296556831e-05,
      "loss": 1.2802,
      "step": 966
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8717512032580527e-05,
      "loss": 1.249,
      "step": 967
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8716031099592745e-05,
      "loss": 1.5309,
      "step": 968
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8714550166604962e-05,
      "loss": 1.4693,
      "step": 969
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.871306923361718e-05,
      "loss": 1.3195,
      "step": 970
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8711588300629397e-05,
      "loss": 1.5796,
      "step": 971
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8710107367641618e-05,
      "loss": 1.4722,
      "step": 972
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8708626434653832e-05,
      "loss": 1.7721,
      "step": 973
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.870714550166605e-05,
      "loss": 1.6493,
      "step": 974
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.8705664568678267e-05,
      "loss": 1.6357,
      "step": 975
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8704183635690488e-05,
      "loss": 1.52,
      "step": 976
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8702702702702706e-05,
      "loss": 1.6383,
      "step": 977
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.870122176971492e-05,
      "loss": 1.5983,
      "step": 978
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.869974083672714e-05,
      "loss": 1.6548,
      "step": 979
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8698259903739358e-05,
      "loss": 1.6464,
      "step": 980
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8696778970751576e-05,
      "loss": 1.5653,
      "step": 981
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8695298037763793e-05,
      "loss": 1.2662,
      "step": 982
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.869381710477601e-05,
      "loss": 1.7478,
      "step": 983
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8692336171788228e-05,
      "loss": 1.4798,
      "step": 984
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8690855238800446e-05,
      "loss": 1.5102,
      "step": 985
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8689374305812663e-05,
      "loss": 1.5815,
      "step": 986
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.868789337282488e-05,
      "loss": 1.5943,
      "step": 987
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.86864124398371e-05,
      "loss": 1.5384,
      "step": 988
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8684931506849316e-05,
      "loss": 1.401,
      "step": 989
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8683450573861533e-05,
      "loss": 1.4694,
      "step": 990
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.868196964087375e-05,
      "loss": 1.4562,
      "step": 991
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.868048870788597e-05,
      "loss": 1.3121,
      "step": 992
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.867900777489819e-05,
      "loss": 1.4341,
      "step": 993
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8677526841910403e-05,
      "loss": 1.5074,
      "step": 994
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.867604590892262e-05,
      "loss": 1.5679,
      "step": 995
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8674564975934842e-05,
      "loss": 1.4588,
      "step": 996
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.867308404294706e-05,
      "loss": 1.5644,
      "step": 997
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8671603109959277e-05,
      "loss": 1.3993,
      "step": 998
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8670122176971494e-05,
      "loss": 1.5342,
      "step": 999
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8668641243983712e-05,
      "loss": 1.544,
      "step": 1000
    },
    {
      "epoch": 0.22,
      "eval_loss": NaN,
      "eval_runtime": 4.3736,
      "eval_samples_per_second": 353.03,
      "eval_steps_per_second": 44.129,
      "step": 1000
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.866716031099593e-05,
      "loss": 1.3128,
      "step": 1001
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8665679378008147e-05,
      "loss": 1.3211,
      "step": 1002
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8664198445020365e-05,
      "loss": 1.8285,
      "step": 1003
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8662717512032582e-05,
      "loss": 1.7618,
      "step": 1004
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.86612365790448e-05,
      "loss": 1.6643,
      "step": 1005
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8659755646057017e-05,
      "loss": 1.6464,
      "step": 1006
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8658274713069235e-05,
      "loss": 1.4973,
      "step": 1007
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8656793780081452e-05,
      "loss": 1.5309,
      "step": 1008
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.865531284709367e-05,
      "loss": 1.2378,
      "step": 1009
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8653831914105887e-05,
      "loss": 1.6876,
      "step": 1010
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8652350981118105e-05,
      "loss": 1.5508,
      "step": 1011
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8650870048130322e-05,
      "loss": 1.4142,
      "step": 1012
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8649389115142543e-05,
      "loss": 1.341,
      "step": 1013
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.864790818215476e-05,
      "loss": 1.1897,
      "step": 1014
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8646427249166975e-05,
      "loss": 1.6249,
      "step": 1015
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8644946316179196e-05,
      "loss": 1.6881,
      "step": 1016
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8643465383191413e-05,
      "loss": 1.4096,
      "step": 1017
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.864198445020363e-05,
      "loss": 1.5742,
      "step": 1018
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8640503517215848e-05,
      "loss": 1.503,
      "step": 1019
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.8639022584228066e-05,
      "loss": 1.5907,
      "step": 1020
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8637541651240283e-05,
      "loss": 1.4842,
      "step": 1021
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.86360607182525e-05,
      "loss": 1.5104,
      "step": 1022
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8634579785264718e-05,
      "loss": 1.4879,
      "step": 1023
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8633098852276936e-05,
      "loss": 1.6358,
      "step": 1024
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8631617919289153e-05,
      "loss": 1.3597,
      "step": 1025
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.863013698630137e-05,
      "loss": 1.5065,
      "step": 1026
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8628656053313588e-05,
      "loss": 1.5986,
      "step": 1027
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8627175120325806e-05,
      "loss": 1.4139,
      "step": 1028
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8625694187338023e-05,
      "loss": 1.5709,
      "step": 1029
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8624213254350244e-05,
      "loss": 1.6411,
      "step": 1030
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.862273232136246e-05,
      "loss": 1.5361,
      "step": 1031
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8621251388374676e-05,
      "loss": 1.4691,
      "step": 1032
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8619770455386897e-05,
      "loss": 1.5395,
      "step": 1033
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8618289522399114e-05,
      "loss": 1.4605,
      "step": 1034
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8616808589411332e-05,
      "loss": 1.5503,
      "step": 1035
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8615327656423546e-05,
      "loss": 1.4513,
      "step": 1036
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8613846723435767e-05,
      "loss": 1.6932,
      "step": 1037
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8612365790447984e-05,
      "loss": 1.4585,
      "step": 1038
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8610884857460202e-05,
      "loss": 1.5689,
      "step": 1039
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.860940392447242e-05,
      "loss": 1.3318,
      "step": 1040
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8607922991484637e-05,
      "loss": 1.5022,
      "step": 1041
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8606442058496854e-05,
      "loss": 1.5371,
      "step": 1042
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8604961125509072e-05,
      "loss": 1.4418,
      "step": 1043
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.860348019252129e-05,
      "loss": 1.6744,
      "step": 1044
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8601999259533507e-05,
      "loss": 1.464,
      "step": 1045
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8600518326545725e-05,
      "loss": 1.5359,
      "step": 1046
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8599037393557945e-05,
      "loss": 1.312,
      "step": 1047
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.859755646057016e-05,
      "loss": 1.624,
      "step": 1048
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8596075527582377e-05,
      "loss": 1.5424,
      "step": 1049
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8594594594594598e-05,
      "loss": 1.5801,
      "step": 1050
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8593113661606815e-05,
      "loss": 1.6169,
      "step": 1051
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.859163272861903e-05,
      "loss": 1.5854,
      "step": 1052
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8590151795631247e-05,
      "loss": 1.5798,
      "step": 1053
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8588670862643468e-05,
      "loss": 1.3837,
      "step": 1054
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8587189929655686e-05,
      "loss": 1.4888,
      "step": 1055
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8585708996667903e-05,
      "loss": 1.7668,
      "step": 1056
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.858422806368012e-05,
      "loss": 1.3702,
      "step": 1057
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8582747130692338e-05,
      "loss": 1.4107,
      "step": 1058
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8581266197704556e-05,
      "loss": 1.4685,
      "step": 1059
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8579785264716773e-05,
      "loss": 1.712,
      "step": 1060
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.857830433172899e-05,
      "loss": 1.2128,
      "step": 1061
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8576823398741208e-05,
      "loss": 1.7331,
      "step": 1062
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8575342465753426e-05,
      "loss": 1.7265,
      "step": 1063
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.8573861532765643e-05,
      "loss": 1.5946,
      "step": 1064
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.857238059977786e-05,
      "loss": 1.5053,
      "step": 1065
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8570899666790078e-05,
      "loss": 1.5188,
      "step": 1066
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.85694187338023e-05,
      "loss": 1.5186,
      "step": 1067
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8567937800814517e-05,
      "loss": 1.5012,
      "step": 1068
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.856645686782673e-05,
      "loss": 1.5014,
      "step": 1069
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8564975934838948e-05,
      "loss": 1.388,
      "step": 1070
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.856349500185117e-05,
      "loss": 1.6183,
      "step": 1071
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8562014068863387e-05,
      "loss": 1.5486,
      "step": 1072
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.85605331358756e-05,
      "loss": 1.4532,
      "step": 1073
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8559052202887822e-05,
      "loss": 1.3842,
      "step": 1074
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.855757126990004e-05,
      "loss": 1.4146,
      "step": 1075
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8556090336912257e-05,
      "loss": 1.4785,
      "step": 1076
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8554609403924474e-05,
      "loss": 1.5089,
      "step": 1077
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8553128470936692e-05,
      "loss": 1.3746,
      "step": 1078
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.855164753794891e-05,
      "loss": 1.4366,
      "step": 1079
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8550166604961127e-05,
      "loss": 1.4693,
      "step": 1080
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8548685671973344e-05,
      "loss": 1.74,
      "step": 1081
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8547204738985562e-05,
      "loss": 1.7514,
      "step": 1082
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.854572380599778e-05,
      "loss": 1.4663,
      "step": 1083
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.854424287301e-05,
      "loss": 1.4618,
      "step": 1084
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8542761940022214e-05,
      "loss": 1.6065,
      "step": 1085
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8541281007034432e-05,
      "loss": 1.5857,
      "step": 1086
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.853980007404665e-05,
      "loss": 1.4198,
      "step": 1087
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.853831914105887e-05,
      "loss": 1.7113,
      "step": 1088
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8536838208071088e-05,
      "loss": 1.5695,
      "step": 1089
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8535357275083302e-05,
      "loss": 1.5686,
      "step": 1090
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8533876342095523e-05,
      "loss": 1.479,
      "step": 1091
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.853239540910774e-05,
      "loss": 1.48,
      "step": 1092
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8530914476119958e-05,
      "loss": 1.638,
      "step": 1093
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8529433543132172e-05,
      "loss": 1.5489,
      "step": 1094
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8527952610144393e-05,
      "loss": 1.684,
      "step": 1095
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.852647167715661e-05,
      "loss": 1.6088,
      "step": 1096
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8524990744168828e-05,
      "loss": 1.4455,
      "step": 1097
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8523509811181046e-05,
      "loss": 1.5974,
      "step": 1098
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8522028878193263e-05,
      "loss": 1.5694,
      "step": 1099
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.852054794520548e-05,
      "loss": 1.6217,
      "step": 1100
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8519067012217698e-05,
      "loss": 1.5603,
      "step": 1101
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8517586079229916e-05,
      "loss": 1.6516,
      "step": 1102
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8516105146242133e-05,
      "loss": 1.418,
      "step": 1103
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.851462421325435e-05,
      "loss": 1.4947,
      "step": 1104
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.851314328026657e-05,
      "loss": 1.424,
      "step": 1105
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8511662347278786e-05,
      "loss": 1.4744,
      "step": 1106
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8510181414291003e-05,
      "loss": 1.6154,
      "step": 1107
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8508700481303224e-05,
      "loss": 1.7471,
      "step": 1108
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.850721954831544e-05,
      "loss": 1.4133,
      "step": 1109
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.850573861532766e-05,
      "loss": 1.4603,
      "step": 1110
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.8504257682339877e-05,
      "loss": 1.7246,
      "step": 1111
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8502776749352094e-05,
      "loss": 1.4675,
      "step": 1112
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8501295816364312e-05,
      "loss": 1.3372,
      "step": 1113
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.849981488337653e-05,
      "loss": 1.4547,
      "step": 1114
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8498333950388747e-05,
      "loss": 1.4005,
      "step": 1115
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8496853017400964e-05,
      "loss": 1.4198,
      "step": 1116
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8495372084413182e-05,
      "loss": 1.4615,
      "step": 1117
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.84938911514254e-05,
      "loss": 1.4647,
      "step": 1118
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8492410218437617e-05,
      "loss": 1.6892,
      "step": 1119
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8490929285449834e-05,
      "loss": 1.4545,
      "step": 1120
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8489448352462052e-05,
      "loss": 1.3173,
      "step": 1121
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.848796741947427e-05,
      "loss": 1.6371,
      "step": 1122
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8486486486486487e-05,
      "loss": 1.3365,
      "step": 1123
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8485005553498704e-05,
      "loss": 1.7392,
      "step": 1124
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8483524620510925e-05,
      "loss": 1.2386,
      "step": 1125
    },
    {
      "epoch": 0.25,
      "eval_loss": NaN,
      "eval_runtime": 4.3777,
      "eval_samples_per_second": 352.697,
      "eval_steps_per_second": 44.087,
      "step": 1125
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8482043687523143e-05,
      "loss": 1.6036,
      "step": 1126
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8480562754535357e-05,
      "loss": 1.35,
      "step": 1127
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8479081821547578e-05,
      "loss": 1.6627,
      "step": 1128
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8477600888559795e-05,
      "loss": 1.4889,
      "step": 1129
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8476119955572013e-05,
      "loss": 1.6049,
      "step": 1130
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.847463902258423e-05,
      "loss": 1.4489,
      "step": 1131
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8473158089596448e-05,
      "loss": 1.6326,
      "step": 1132
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8471677156608665e-05,
      "loss": 1.63,
      "step": 1133
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8470196223620883e-05,
      "loss": 1.535,
      "step": 1134
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.84687152906331e-05,
      "loss": 1.2777,
      "step": 1135
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8467234357645318e-05,
      "loss": 1.5656,
      "step": 1136
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8465753424657535e-05,
      "loss": 1.7281,
      "step": 1137
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8464272491669753e-05,
      "loss": 1.4038,
      "step": 1138
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.846279155868197e-05,
      "loss": 1.5849,
      "step": 1139
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8461310625694188e-05,
      "loss": 1.38,
      "step": 1140
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8459829692706406e-05,
      "loss": 1.6095,
      "step": 1141
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8458348759718626e-05,
      "loss": 1.5049,
      "step": 1142
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.845686782673084e-05,
      "loss": 1.6219,
      "step": 1143
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8455386893743058e-05,
      "loss": 1.5954,
      "step": 1144
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.845390596075528e-05,
      "loss": 1.6185,
      "step": 1145
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8452425027767497e-05,
      "loss": 1.3894,
      "step": 1146
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8450944094779714e-05,
      "loss": 1.7488,
      "step": 1147
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8449463161791928e-05,
      "loss": 1.5729,
      "step": 1148
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.844798222880415e-05,
      "loss": 1.4165,
      "step": 1149
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8446501295816367e-05,
      "loss": 1.5335,
      "step": 1150
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8445020362828584e-05,
      "loss": 1.294,
      "step": 1151
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.84435394298408e-05,
      "loss": 1.443,
      "step": 1152
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.844205849685302e-05,
      "loss": 1.5948,
      "step": 1153
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8440577563865237e-05,
      "loss": 1.5699,
      "step": 1154
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8439096630877454e-05,
      "loss": 1.4719,
      "step": 1155
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.8437615697889672e-05,
      "loss": 1.7296,
      "step": 1156
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.843613476490189e-05,
      "loss": 1.7662,
      "step": 1157
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8434653831914107e-05,
      "loss": 1.256,
      "step": 1158
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8433172898926324e-05,
      "loss": 1.562,
      "step": 1159
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8431691965938542e-05,
      "loss": 1.5379,
      "step": 1160
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.843021103295076e-05,
      "loss": 1.474,
      "step": 1161
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.842873009996298e-05,
      "loss": 1.5788,
      "step": 1162
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8427249166975198e-05,
      "loss": 1.6399,
      "step": 1163
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8425768233987412e-05,
      "loss": 1.5753,
      "step": 1164
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.842428730099963e-05,
      "loss": 1.3609,
      "step": 1165
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.842280636801185e-05,
      "loss": 1.668,
      "step": 1166
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8421325435024068e-05,
      "loss": 1.3774,
      "step": 1167
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8419844502036285e-05,
      "loss": 1.5342,
      "step": 1168
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8418363569048503e-05,
      "loss": 1.8934,
      "step": 1169
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.841688263606072e-05,
      "loss": 1.5754,
      "step": 1170
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8415401703072938e-05,
      "loss": 1.6296,
      "step": 1171
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8413920770085155e-05,
      "loss": 1.412,
      "step": 1172
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8412439837097373e-05,
      "loss": 1.3476,
      "step": 1173
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.841095890410959e-05,
      "loss": 1.5705,
      "step": 1174
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8409477971121808e-05,
      "loss": 1.6373,
      "step": 1175
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8407997038134025e-05,
      "loss": 1.5453,
      "step": 1176
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8406516105146243e-05,
      "loss": 1.2538,
      "step": 1177
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.840503517215846e-05,
      "loss": 1.3394,
      "step": 1178
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.840355423917068e-05,
      "loss": 1.5211,
      "step": 1179
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8402073306182895e-05,
      "loss": 1.5136,
      "step": 1180
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8400592373195113e-05,
      "loss": 1.5937,
      "step": 1181
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.839911144020733e-05,
      "loss": 1.5845,
      "step": 1182
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.839763050721955e-05,
      "loss": 1.4827,
      "step": 1183
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.839614957423177e-05,
      "loss": 1.6074,
      "step": 1184
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8394668641243983e-05,
      "loss": 1.4609,
      "step": 1185
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8393187708256204e-05,
      "loss": 1.2886,
      "step": 1186
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.839170677526842e-05,
      "loss": 1.4292,
      "step": 1187
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.839022584228064e-05,
      "loss": 1.539,
      "step": 1188
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8388744909292857e-05,
      "loss": 1.4746,
      "step": 1189
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8387263976305074e-05,
      "loss": 1.3304,
      "step": 1190
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.838578304331729e-05,
      "loss": 1.5153,
      "step": 1191
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.838430211032951e-05,
      "loss": 1.5138,
      "step": 1192
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8382821177341727e-05,
      "loss": 1.668,
      "step": 1193
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8381340244353944e-05,
      "loss": 1.5529,
      "step": 1194
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.837985931136616e-05,
      "loss": 1.2946,
      "step": 1195
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8378378378378383e-05,
      "loss": 1.5269,
      "step": 1196
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8376897445390597e-05,
      "loss": 1.4212,
      "step": 1197
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8375416512402814e-05,
      "loss": 1.6,
      "step": 1198
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8373935579415032e-05,
      "loss": 1.4515,
      "step": 1199
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8372454646427253e-05,
      "loss": 1.4777,
      "step": 1200
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8370973713439467e-05,
      "loss": 1.4576,
      "step": 1201
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8369492780451684e-05,
      "loss": 1.2351,
      "step": 1202
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8368011847463905e-05,
      "loss": 1.6153,
      "step": 1203
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8366530914476123e-05,
      "loss": 1.3511,
      "step": 1204
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.836504998148834e-05,
      "loss": 1.5531,
      "step": 1205
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8363569048500554e-05,
      "loss": 1.4371,
      "step": 1206
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8362088115512775e-05,
      "loss": 1.1996,
      "step": 1207
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8360607182524993e-05,
      "loss": 1.7561,
      "step": 1208
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.835912624953721e-05,
      "loss": 1.7475,
      "step": 1209
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8357645316549428e-05,
      "loss": 1.5481,
      "step": 1210
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8356164383561645e-05,
      "loss": 1.7038,
      "step": 1211
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8354683450573863e-05,
      "loss": 1.5431,
      "step": 1212
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.835320251758608e-05,
      "loss": 1.3114,
      "step": 1213
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8351721584598298e-05,
      "loss": 1.3826,
      "step": 1214
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8350240651610515e-05,
      "loss": 1.5254,
      "step": 1215
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8348759718622733e-05,
      "loss": 1.6211,
      "step": 1216
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8347278785634954e-05,
      "loss": 1.5108,
      "step": 1217
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8345797852647168e-05,
      "loss": 1.4326,
      "step": 1218
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8344316919659385e-05,
      "loss": 1.7356,
      "step": 1219
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8342835986671606e-05,
      "loss": 1.7382,
      "step": 1220
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8341355053683824e-05,
      "loss": 1.5093,
      "step": 1221
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8339874120696038e-05,
      "loss": 1.7869,
      "step": 1222
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8338393187708255e-05,
      "loss": 1.681,
      "step": 1223
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8336912254720476e-05,
      "loss": 1.6606,
      "step": 1224
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8335431321732694e-05,
      "loss": 1.4754,
      "step": 1225
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.833395038874491e-05,
      "loss": 1.6958,
      "step": 1226
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.833246945575713e-05,
      "loss": 1.4912,
      "step": 1227
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8330988522769346e-05,
      "loss": 1.2973,
      "step": 1228
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8329507589781564e-05,
      "loss": 1.4819,
      "step": 1229
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.832802665679378e-05,
      "loss": 1.4803,
      "step": 1230
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8326545723806e-05,
      "loss": 1.5312,
      "step": 1231
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8325064790818217e-05,
      "loss": 1.4888,
      "step": 1232
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8323583857830434e-05,
      "loss": 1.5329,
      "step": 1233
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.832210292484265e-05,
      "loss": 1.4243,
      "step": 1234
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.832062199185487e-05,
      "loss": 1.6433,
      "step": 1235
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8319141058867087e-05,
      "loss": 1.3971,
      "step": 1236
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8317660125879308e-05,
      "loss": 1.4284,
      "step": 1237
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8316179192891525e-05,
      "loss": 1.4137,
      "step": 1238
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.831469825990374e-05,
      "loss": 1.5732,
      "step": 1239
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.831321732691596e-05,
      "loss": 1.5366,
      "step": 1240
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8311736393928178e-05,
      "loss": 1.4947,
      "step": 1241
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8310255460940395e-05,
      "loss": 1.5515,
      "step": 1242
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.830877452795261e-05,
      "loss": 1.5074,
      "step": 1243
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.830729359496483e-05,
      "loss": 1.6739,
      "step": 1244
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8305812661977048e-05,
      "loss": 1.4317,
      "step": 1245
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8304331728989265e-05,
      "loss": 1.6173,
      "step": 1246
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.8302850796001483e-05,
      "loss": 1.5012,
      "step": 1247
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.83013698630137e-05,
      "loss": 1.7872,
      "step": 1248
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8299888930025918e-05,
      "loss": 1.5121,
      "step": 1249
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8298407997038135e-05,
      "loss": 1.5908,
      "step": 1250
    },
    {
      "epoch": 0.28,
      "eval_loss": NaN,
      "eval_runtime": 4.373,
      "eval_samples_per_second": 353.074,
      "eval_steps_per_second": 44.134,
      "step": 1250
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8296927064050353e-05,
      "loss": 1.8195,
      "step": 1251
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.829544613106257e-05,
      "loss": 1.5463,
      "step": 1252
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8293965198074788e-05,
      "loss": 1.355,
      "step": 1253
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.829248426508701e-05,
      "loss": 1.2953,
      "step": 1254
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8291003332099223e-05,
      "loss": 1.5076,
      "step": 1255
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.828952239911144e-05,
      "loss": 1.5762,
      "step": 1256
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.828804146612366e-05,
      "loss": 1.5098,
      "step": 1257
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.828656053313588e-05,
      "loss": 1.5022,
      "step": 1258
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8285079600148096e-05,
      "loss": 1.4456,
      "step": 1259
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.828359866716031e-05,
      "loss": 1.491,
      "step": 1260
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.828211773417253e-05,
      "loss": 1.7865,
      "step": 1261
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.828063680118475e-05,
      "loss": 1.6076,
      "step": 1262
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8279155868196966e-05,
      "loss": 1.4311,
      "step": 1263
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8277674935209184e-05,
      "loss": 1.7037,
      "step": 1264
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.82761940022214e-05,
      "loss": 1.6489,
      "step": 1265
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.827471306923362e-05,
      "loss": 1.5753,
      "step": 1266
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8273232136245836e-05,
      "loss": 1.4811,
      "step": 1267
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8271751203258054e-05,
      "loss": 1.5854,
      "step": 1268
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.827027027027027e-05,
      "loss": 1.3491,
      "step": 1269
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.826878933728249e-05,
      "loss": 1.6741,
      "step": 1270
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8267308404294706e-05,
      "loss": 1.4358,
      "step": 1271
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8265827471306924e-05,
      "loss": 1.2093,
      "step": 1272
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.826434653831914e-05,
      "loss": 1.6055,
      "step": 1273
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8262865605331362e-05,
      "loss": 1.4602,
      "step": 1274
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.826138467234358e-05,
      "loss": 1.4381,
      "step": 1275
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8259903739355794e-05,
      "loss": 1.5247,
      "step": 1276
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.825842280636801e-05,
      "loss": 1.4797,
      "step": 1277
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8256941873380232e-05,
      "loss": 1.5117,
      "step": 1278
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.825546094039245e-05,
      "loss": 1.4131,
      "step": 1279
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8253980007404664e-05,
      "loss": 1.4464,
      "step": 1280
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8252499074416885e-05,
      "loss": 1.4151,
      "step": 1281
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8251018141429103e-05,
      "loss": 1.4446,
      "step": 1282
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.824953720844132e-05,
      "loss": 1.3863,
      "step": 1283
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8248056275453538e-05,
      "loss": 1.5126,
      "step": 1284
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8246575342465755e-05,
      "loss": 1.5277,
      "step": 1285
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8245094409477973e-05,
      "loss": 1.4867,
      "step": 1286
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.824361347649019e-05,
      "loss": 1.4674,
      "step": 1287
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8242132543502408e-05,
      "loss": 1.4336,
      "step": 1288
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8240651610514625e-05,
      "loss": 1.2924,
      "step": 1289
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8239170677526843e-05,
      "loss": 1.3971,
      "step": 1290
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8237689744539064e-05,
      "loss": 1.5183,
      "step": 1291
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8236208811551278e-05,
      "loss": 1.4498,
      "step": 1292
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8234727878563495e-05,
      "loss": 1.3541,
      "step": 1293
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8233246945575713e-05,
      "loss": 1.5106,
      "step": 1294
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8231766012587934e-05,
      "loss": 1.6447,
      "step": 1295
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.823028507960015e-05,
      "loss": 1.6784,
      "step": 1296
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8228804146612365e-05,
      "loss": 1.5486,
      "step": 1297
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8227323213624586e-05,
      "loss": 1.6114,
      "step": 1298
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8225842280636804e-05,
      "loss": 1.3573,
      "step": 1299
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.822436134764902e-05,
      "loss": 1.5388,
      "step": 1300
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8222880414661235e-05,
      "loss": 1.5565,
      "step": 1301
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8221399481673456e-05,
      "loss": 1.5282,
      "step": 1302
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8219918548685674e-05,
      "loss": 1.4813,
      "step": 1303
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.821843761569789e-05,
      "loss": 1.8961,
      "step": 1304
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.821695668271011e-05,
      "loss": 1.4578,
      "step": 1305
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8215475749722326e-05,
      "loss": 1.6508,
      "step": 1306
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8213994816734544e-05,
      "loss": 1.5291,
      "step": 1307
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.821251388374676e-05,
      "loss": 1.2622,
      "step": 1308
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.821103295075898e-05,
      "loss": 1.5133,
      "step": 1309
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8209552017771196e-05,
      "loss": 1.4662,
      "step": 1310
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8208071084783414e-05,
      "loss": 1.6146,
      "step": 1311
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8206590151795635e-05,
      "loss": 1.4332,
      "step": 1312
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.820510921880785e-05,
      "loss": 1.6706,
      "step": 1313
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8203628285820066e-05,
      "loss": 1.2916,
      "step": 1314
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8202147352832287e-05,
      "loss": 1.4655,
      "step": 1315
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8200666419844505e-05,
      "loss": 1.4496,
      "step": 1316
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8199185486856722e-05,
      "loss": 1.4329,
      "step": 1317
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8197704553868937e-05,
      "loss": 1.6143,
      "step": 1318
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8196223620881157e-05,
      "loss": 1.5975,
      "step": 1319
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8194742687893375e-05,
      "loss": 1.4088,
      "step": 1320
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8193261754905592e-05,
      "loss": 1.3003,
      "step": 1321
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.819178082191781e-05,
      "loss": 1.4109,
      "step": 1322
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8190299888930028e-05,
      "loss": 1.6367,
      "step": 1323
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8188818955942245e-05,
      "loss": 1.8282,
      "step": 1324
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8187338022954463e-05,
      "loss": 1.5474,
      "step": 1325
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.818585708996668e-05,
      "loss": 1.7281,
      "step": 1326
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8184376156978898e-05,
      "loss": 1.5089,
      "step": 1327
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8182895223991115e-05,
      "loss": 1.5365,
      "step": 1328
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8181414291003333e-05,
      "loss": 1.3713,
      "step": 1329
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.817993335801555e-05,
      "loss": 1.3798,
      "step": 1330
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8178452425027768e-05,
      "loss": 1.3512,
      "step": 1331
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.817697149203999e-05,
      "loss": 1.3826,
      "step": 1332
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8175490559052206e-05,
      "loss": 1.8247,
      "step": 1333
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.817400962606442e-05,
      "loss": 1.4953,
      "step": 1334
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8172528693076638e-05,
      "loss": 1.3968,
      "step": 1335
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.817104776008886e-05,
      "loss": 1.5705,
      "step": 1336
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.8169566827101076e-05,
      "loss": 1.2437,
      "step": 1337
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8168085894113294e-05,
      "loss": 1.5942,
      "step": 1338
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.816660496112551e-05,
      "loss": 1.4587,
      "step": 1339
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.816512402813773e-05,
      "loss": 1.4863,
      "step": 1340
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8163643095149946e-05,
      "loss": 1.609,
      "step": 1341
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8162162162162164e-05,
      "loss": 1.3973,
      "step": 1342
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.816068122917438e-05,
      "loss": 1.6544,
      "step": 1343
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.81592002961866e-05,
      "loss": 1.479,
      "step": 1344
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8157719363198816e-05,
      "loss": 1.5098,
      "step": 1345
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8156238430211034e-05,
      "loss": 1.6968,
      "step": 1346
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.815475749722325e-05,
      "loss": 1.4662,
      "step": 1347
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.815327656423547e-05,
      "loss": 1.3827,
      "step": 1348
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.815179563124769e-05,
      "loss": 1.517,
      "step": 1349
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8150314698259904e-05,
      "loss": 1.5885,
      "step": 1350
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.814883376527212e-05,
      "loss": 1.4456,
      "step": 1351
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.814735283228434e-05,
      "loss": 1.7054,
      "step": 1352
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.814587189929656e-05,
      "loss": 1.568,
      "step": 1353
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8144390966308777e-05,
      "loss": 1.4673,
      "step": 1354
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.814291003332099e-05,
      "loss": 1.5367,
      "step": 1355
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8141429100333212e-05,
      "loss": 1.5448,
      "step": 1356
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.813994816734543e-05,
      "loss": 1.4603,
      "step": 1357
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8138467234357647e-05,
      "loss": 1.5538,
      "step": 1358
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8136986301369865e-05,
      "loss": 1.5744,
      "step": 1359
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8135505368382082e-05,
      "loss": 1.368,
      "step": 1360
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.81340244353943e-05,
      "loss": 1.3154,
      "step": 1361
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8132543502406517e-05,
      "loss": 1.2725,
      "step": 1362
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8131062569418735e-05,
      "loss": 1.6407,
      "step": 1363
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8129581636430952e-05,
      "loss": 1.3186,
      "step": 1364
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.812810070344317e-05,
      "loss": 1.5644,
      "step": 1365
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.812661977045539e-05,
      "loss": 1.4416,
      "step": 1366
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8125138837467605e-05,
      "loss": 1.6657,
      "step": 1367
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8123657904479823e-05,
      "loss": 1.4116,
      "step": 1368
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.812217697149204e-05,
      "loss": 1.3782,
      "step": 1369
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.812069603850426e-05,
      "loss": 1.6535,
      "step": 1370
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8119215105516475e-05,
      "loss": 1.6107,
      "step": 1371
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8117734172528693e-05,
      "loss": 1.4944,
      "step": 1372
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8116253239540914e-05,
      "loss": 1.3233,
      "step": 1373
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.811477230655313e-05,
      "loss": 1.6013,
      "step": 1374
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.811329137356535e-05,
      "loss": 1.4459,
      "step": 1375
    },
    {
      "epoch": 0.3,
      "eval_loss": NaN,
      "eval_runtime": 4.378,
      "eval_samples_per_second": 352.673,
      "eval_steps_per_second": 44.084,
      "step": 1375
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8111810440577566e-05,
      "loss": 1.7171,
      "step": 1376
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8110329507589784e-05,
      "loss": 1.6097,
      "step": 1377
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8108848574602e-05,
      "loss": 1.5825,
      "step": 1378
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.810736764161422e-05,
      "loss": 1.2905,
      "step": 1379
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8105886708626436e-05,
      "loss": 1.4307,
      "step": 1380
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8104405775638654e-05,
      "loss": 1.7108,
      "step": 1381
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.810292484265087e-05,
      "loss": 1.7621,
      "step": 1382
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.810144390966309e-05,
      "loss": 1.5487,
      "step": 1383
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8099962976675306e-05,
      "loss": 1.2403,
      "step": 1384
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8098482043687524e-05,
      "loss": 1.5671,
      "step": 1385
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8097001110699745e-05,
      "loss": 1.269,
      "step": 1386
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8095520177711962e-05,
      "loss": 1.452,
      "step": 1387
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8094039244724176e-05,
      "loss": 1.5815,
      "step": 1388
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8092558311736394e-05,
      "loss": 1.587,
      "step": 1389
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8091077378748615e-05,
      "loss": 1.5278,
      "step": 1390
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8089596445760832e-05,
      "loss": 1.2768,
      "step": 1391
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8088115512773046e-05,
      "loss": 1.3384,
      "step": 1392
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8086634579785267e-05,
      "loss": 1.5323,
      "step": 1393
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8085153646797485e-05,
      "loss": 1.4378,
      "step": 1394
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8083672713809702e-05,
      "loss": 1.2764,
      "step": 1395
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.808219178082192e-05,
      "loss": 1.441,
      "step": 1396
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8080710847834137e-05,
      "loss": 1.2468,
      "step": 1397
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8079229914846355e-05,
      "loss": 1.6,
      "step": 1398
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8077748981858572e-05,
      "loss": 1.4569,
      "step": 1399
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.807626804887079e-05,
      "loss": 1.5068,
      "step": 1400
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8074787115883007e-05,
      "loss": 1.9199,
      "step": 1401
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8073306182895225e-05,
      "loss": 1.3411,
      "step": 1402
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8071825249907446e-05,
      "loss": 1.4017,
      "step": 1403
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.807034431691966e-05,
      "loss": 1.3902,
      "step": 1404
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8068863383931877e-05,
      "loss": 1.3809,
      "step": 1405
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8067382450944095e-05,
      "loss": 1.4601,
      "step": 1406
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8065901517956316e-05,
      "loss": 1.4934,
      "step": 1407
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.806442058496853e-05,
      "loss": 1.812,
      "step": 1408
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8062939651980748e-05,
      "loss": 1.3651,
      "step": 1409
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.806145871899297e-05,
      "loss": 1.5584,
      "step": 1410
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8059977786005186e-05,
      "loss": 1.3941,
      "step": 1411
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8058496853017403e-05,
      "loss": 1.365,
      "step": 1412
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8057015920029618e-05,
      "loss": 1.282,
      "step": 1413
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.805553498704184e-05,
      "loss": 1.4371,
      "step": 1414
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8054054054054056e-05,
      "loss": 1.3283,
      "step": 1415
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8052573121066274e-05,
      "loss": 1.4922,
      "step": 1416
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.805109218807849e-05,
      "loss": 1.4236,
      "step": 1417
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.804961125509071e-05,
      "loss": 1.3881,
      "step": 1418
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8048130322102926e-05,
      "loss": 1.7617,
      "step": 1419
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8046649389115144e-05,
      "loss": 1.4862,
      "step": 1420
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.804516845612736e-05,
      "loss": 1.5471,
      "step": 1421
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.804368752313958e-05,
      "loss": 2.0483,
      "step": 1422
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8042206590151796e-05,
      "loss": 1.4193,
      "step": 1423
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8040725657164017e-05,
      "loss": 1.482,
      "step": 1424
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.803924472417623e-05,
      "loss": 1.623,
      "step": 1425
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.803776379118845e-05,
      "loss": 1.417,
      "step": 1426
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.803628285820067e-05,
      "loss": 1.6279,
      "step": 1427
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.8034801925212887e-05,
      "loss": 1.5604,
      "step": 1428
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.80333209922251e-05,
      "loss": 1.4438,
      "step": 1429
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.803184005923732e-05,
      "loss": 1.442,
      "step": 1430
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.803035912624954e-05,
      "loss": 1.2784,
      "step": 1431
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8028878193261757e-05,
      "loss": 1.4069,
      "step": 1432
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8027397260273975e-05,
      "loss": 1.4113,
      "step": 1433
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8025916327286192e-05,
      "loss": 1.2224,
      "step": 1434
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.802443539429841e-05,
      "loss": 1.4245,
      "step": 1435
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8022954461310627e-05,
      "loss": 1.6236,
      "step": 1436
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8021473528322845e-05,
      "loss": 1.3755,
      "step": 1437
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8019992595335062e-05,
      "loss": 1.3421,
      "step": 1438
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.801851166234728e-05,
      "loss": 1.5576,
      "step": 1439
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8017030729359497e-05,
      "loss": 1.656,
      "step": 1440
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8015549796371715e-05,
      "loss": 1.3042,
      "step": 1441
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8014068863383932e-05,
      "loss": 1.294,
      "step": 1442
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.801258793039615e-05,
      "loss": 1.6061,
      "step": 1443
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.801110699740837e-05,
      "loss": 1.58,
      "step": 1444
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8009626064420588e-05,
      "loss": 1.6866,
      "step": 1445
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8008145131432802e-05,
      "loss": 1.6351,
      "step": 1446
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.800666419844502e-05,
      "loss": 1.4837,
      "step": 1447
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.800518326545724e-05,
      "loss": 1.7878,
      "step": 1448
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.800370233246946e-05,
      "loss": 1.6414,
      "step": 1449
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8002221399481672e-05,
      "loss": 1.4324,
      "step": 1450
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8000740466493893e-05,
      "loss": 1.3973,
      "step": 1451
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.799925953350611e-05,
      "loss": 1.5935,
      "step": 1452
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.799777860051833e-05,
      "loss": 1.4097,
      "step": 1453
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7996297667530546e-05,
      "loss": 1.2737,
      "step": 1454
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7994816734542763e-05,
      "loss": 1.5236,
      "step": 1455
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.799333580155498e-05,
      "loss": 1.2921,
      "step": 1456
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.79918548685672e-05,
      "loss": 1.5875,
      "step": 1457
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7990373935579416e-05,
      "loss": 1.4008,
      "step": 1458
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7988893002591634e-05,
      "loss": 1.4229,
      "step": 1459
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.798741206960385e-05,
      "loss": 1.6041,
      "step": 1460
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7985931136616072e-05,
      "loss": 1.5789,
      "step": 1461
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7984450203628286e-05,
      "loss": 1.7135,
      "step": 1462
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7982969270640504e-05,
      "loss": 1.4521,
      "step": 1463
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.798148833765272e-05,
      "loss": 1.6438,
      "step": 1464
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7980007404664942e-05,
      "loss": 1.3254,
      "step": 1465
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.797852647167716e-05,
      "loss": 1.3343,
      "step": 1466
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7977045538689374e-05,
      "loss": 1.5282,
      "step": 1467
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7975564605701595e-05,
      "loss": 1.6722,
      "step": 1468
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7974083672713812e-05,
      "loss": 1.6529,
      "step": 1469
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.797260273972603e-05,
      "loss": 1.4545,
      "step": 1470
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7971121806738247e-05,
      "loss": 1.4885,
      "step": 1471
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7969640873750465e-05,
      "loss": 1.6417,
      "step": 1472
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.7968159940762682e-05,
      "loss": 1.4307,
      "step": 1473
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.79666790077749e-05,
      "loss": 1.7021,
      "step": 1474
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7965198074787117e-05,
      "loss": 1.514,
      "step": 1475
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7963717141799335e-05,
      "loss": 1.6856,
      "step": 1476
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7962236208811552e-05,
      "loss": 1.607,
      "step": 1477
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.796075527582377e-05,
      "loss": 1.6934,
      "step": 1478
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7959274342835987e-05,
      "loss": 1.4468,
      "step": 1479
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7957793409848205e-05,
      "loss": 1.3222,
      "step": 1480
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7956312476860422e-05,
      "loss": 1.4356,
      "step": 1481
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7954831543872643e-05,
      "loss": 1.4904,
      "step": 1482
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7953350610884857e-05,
      "loss": 1.4053,
      "step": 1483
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7951869677897075e-05,
      "loss": 1.7084,
      "step": 1484
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7950388744909296e-05,
      "loss": 1.47,
      "step": 1485
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7948907811921513e-05,
      "loss": 1.3327,
      "step": 1486
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.794742687893373e-05,
      "loss": 1.5754,
      "step": 1487
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7945945945945948e-05,
      "loss": 1.4613,
      "step": 1488
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7944465012958166e-05,
      "loss": 1.572,
      "step": 1489
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7942984079970383e-05,
      "loss": 1.595,
      "step": 1490
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.79415031469826e-05,
      "loss": 1.5709,
      "step": 1491
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.794002221399482e-05,
      "loss": 1.3126,
      "step": 1492
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7938541281007036e-05,
      "loss": 1.4401,
      "step": 1493
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7937060348019253e-05,
      "loss": 1.5682,
      "step": 1494
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.793557941503147e-05,
      "loss": 1.4821,
      "step": 1495
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.793409848204369e-05,
      "loss": 1.6108,
      "step": 1496
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7932617549055906e-05,
      "loss": 1.4955,
      "step": 1497
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7931136616068123e-05,
      "loss": 1.4434,
      "step": 1498
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.792965568308034e-05,
      "loss": 1.5926,
      "step": 1499
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.792817475009256e-05,
      "loss": 1.6012,
      "step": 1500
    },
    {
      "epoch": 0.33,
      "eval_loss": NaN,
      "eval_runtime": 4.3732,
      "eval_samples_per_second": 353.062,
      "eval_steps_per_second": 44.133,
      "step": 1500
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7926693817104776e-05,
      "loss": 1.6276,
      "step": 1501
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7925212884116997e-05,
      "loss": 1.6101,
      "step": 1502
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7923731951129214e-05,
      "loss": 1.4472,
      "step": 1503
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.792225101814143e-05,
      "loss": 1.3061,
      "step": 1504
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.792077008515365e-05,
      "loss": 1.4804,
      "step": 1505
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7919289152165867e-05,
      "loss": 1.3197,
      "step": 1506
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7917808219178085e-05,
      "loss": 1.6706,
      "step": 1507
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7916327286190302e-05,
      "loss": 1.4031,
      "step": 1508
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.791484635320252e-05,
      "loss": 1.6638,
      "step": 1509
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7913365420214737e-05,
      "loss": 1.4643,
      "step": 1510
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7911884487226955e-05,
      "loss": 1.5223,
      "step": 1511
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7910403554239172e-05,
      "loss": 1.5317,
      "step": 1512
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.790892262125139e-05,
      "loss": 1.4319,
      "step": 1513
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7907441688263607e-05,
      "loss": 1.5766,
      "step": 1514
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7905960755275828e-05,
      "loss": 1.7772,
      "step": 1515
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7904479822288042e-05,
      "loss": 1.836,
      "step": 1516
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.790299888930026e-05,
      "loss": 1.6248,
      "step": 1517
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7901517956312477e-05,
      "loss": 1.7926,
      "step": 1518
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.7900037023324698e-05,
      "loss": 1.476,
      "step": 1519
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7898556090336912e-05,
      "loss": 1.6825,
      "step": 1520
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.789707515734913e-05,
      "loss": 1.4598,
      "step": 1521
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.789559422436135e-05,
      "loss": 1.6265,
      "step": 1522
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7894113291373568e-05,
      "loss": 1.3182,
      "step": 1523
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7892632358385786e-05,
      "loss": 1.2737,
      "step": 1524
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7891151425398e-05,
      "loss": 1.4657,
      "step": 1525
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.788967049241022e-05,
      "loss": 1.5513,
      "step": 1526
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7888189559422438e-05,
      "loss": 1.532,
      "step": 1527
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7886708626434656e-05,
      "loss": 1.4815,
      "step": 1528
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7885227693446873e-05,
      "loss": 1.5252,
      "step": 1529
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.788374676045909e-05,
      "loss": 1.4066,
      "step": 1530
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7882265827471308e-05,
      "loss": 1.3036,
      "step": 1531
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7880784894483526e-05,
      "loss": 1.3706,
      "step": 1532
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7879303961495743e-05,
      "loss": 1.4842,
      "step": 1533
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.787782302850796e-05,
      "loss": 1.6725,
      "step": 1534
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.787634209552018e-05,
      "loss": 1.3813,
      "step": 1535
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7874861162532396e-05,
      "loss": 1.5241,
      "step": 1536
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7873380229544613e-05,
      "loss": 1.3686,
      "step": 1537
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.787189929655683e-05,
      "loss": 1.6449,
      "step": 1538
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7870418363569052e-05,
      "loss": 1.2846,
      "step": 1539
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.786893743058127e-05,
      "loss": 1.7236,
      "step": 1540
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7867456497593483e-05,
      "loss": 1.5036,
      "step": 1541
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.78659755646057e-05,
      "loss": 1.2414,
      "step": 1542
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7864494631617922e-05,
      "loss": 1.4052,
      "step": 1543
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.786301369863014e-05,
      "loss": 1.5497,
      "step": 1544
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7861532765642357e-05,
      "loss": 1.5298,
      "step": 1545
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7860051832654574e-05,
      "loss": 1.2913,
      "step": 1546
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7858570899666792e-05,
      "loss": 1.4327,
      "step": 1547
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.785708996667901e-05,
      "loss": 1.4184,
      "step": 1548
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7855609033691227e-05,
      "loss": 1.637,
      "step": 1549
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7854128100703445e-05,
      "loss": 1.6173,
      "step": 1550
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7852647167715662e-05,
      "loss": 1.4355,
      "step": 1551
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.785116623472788e-05,
      "loss": 1.5121,
      "step": 1552
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7849685301740097e-05,
      "loss": 1.8282,
      "step": 1553
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7848204368752315e-05,
      "loss": 1.3523,
      "step": 1554
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7846723435764532e-05,
      "loss": 1.4558,
      "step": 1555
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7845242502776753e-05,
      "loss": 1.3414,
      "step": 1556
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7843761569788967e-05,
      "loss": 1.5809,
      "step": 1557
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7842280636801185e-05,
      "loss": 1.6189,
      "step": 1558
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7840799703813402e-05,
      "loss": 1.3698,
      "step": 1559
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7839318770825623e-05,
      "loss": 1.525,
      "step": 1560
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.783783783783784e-05,
      "loss": 1.5705,
      "step": 1561
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7836356904850055e-05,
      "loss": 1.2116,
      "step": 1562
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7834875971862276e-05,
      "loss": 1.3949,
      "step": 1563
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.7833395038874493e-05,
      "loss": 1.4901,
      "step": 1564
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.783191410588671e-05,
      "loss": 1.4481,
      "step": 1565
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7830433172898928e-05,
      "loss": 1.4298,
      "step": 1566
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7828952239911146e-05,
      "loss": 1.7,
      "step": 1567
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7827471306923363e-05,
      "loss": 1.5154,
      "step": 1568
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.782599037393558e-05,
      "loss": 1.5826,
      "step": 1569
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7824509440947798e-05,
      "loss": 1.4965,
      "step": 1570
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7823028507960016e-05,
      "loss": 1.4546,
      "step": 1571
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7821547574972233e-05,
      "loss": 1.5269,
      "step": 1572
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7820066641984454e-05,
      "loss": 1.4632,
      "step": 1573
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7818585708996668e-05,
      "loss": 1.4821,
      "step": 1574
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7817104776008886e-05,
      "loss": 1.5786,
      "step": 1575
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7815623843021103e-05,
      "loss": 1.4637,
      "step": 1576
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7814142910033324e-05,
      "loss": 1.4649,
      "step": 1577
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.781266197704554e-05,
      "loss": 1.2525,
      "step": 1578
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7811181044057756e-05,
      "loss": 1.3363,
      "step": 1579
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7809700111069977e-05,
      "loss": 1.517,
      "step": 1580
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7808219178082194e-05,
      "loss": 1.4479,
      "step": 1581
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7806738245094412e-05,
      "loss": 1.2527,
      "step": 1582
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.780525731210663e-05,
      "loss": 1.5518,
      "step": 1583
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7803776379118847e-05,
      "loss": 1.4333,
      "step": 1584
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7802295446131064e-05,
      "loss": 1.4669,
      "step": 1585
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7800814513143282e-05,
      "loss": 1.5751,
      "step": 1586
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.77993335801555e-05,
      "loss": 1.7484,
      "step": 1587
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7797852647167717e-05,
      "loss": 1.3379,
      "step": 1588
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7796371714179934e-05,
      "loss": 1.531,
      "step": 1589
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7794890781192152e-05,
      "loss": 1.4685,
      "step": 1590
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.779340984820437e-05,
      "loss": 1.632,
      "step": 1591
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7791928915216587e-05,
      "loss": 1.4716,
      "step": 1592
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7790447982228805e-05,
      "loss": 1.424,
      "step": 1593
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7788967049241025e-05,
      "loss": 1.418,
      "step": 1594
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.778748611625324e-05,
      "loss": 1.4089,
      "step": 1595
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7786005183265457e-05,
      "loss": 1.5428,
      "step": 1596
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7784524250277678e-05,
      "loss": 1.6805,
      "step": 1597
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7783043317289895e-05,
      "loss": 1.3216,
      "step": 1598
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.778156238430211e-05,
      "loss": 1.6047,
      "step": 1599
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.778008145131433e-05,
      "loss": 1.5059,
      "step": 1600
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7778600518326548e-05,
      "loss": 1.2831,
      "step": 1601
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7777119585338766e-05,
      "loss": 1.6337,
      "step": 1602
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7775638652350983e-05,
      "loss": 1.64,
      "step": 1603
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.77741577193632e-05,
      "loss": 1.4898,
      "step": 1604
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7772676786375418e-05,
      "loss": 1.3471,
      "step": 1605
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7771195853387636e-05,
      "loss": 1.4388,
      "step": 1606
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7769714920399853e-05,
      "loss": 1.4245,
      "step": 1607
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.776823398741207e-05,
      "loss": 1.613,
      "step": 1608
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.7766753054424288e-05,
      "loss": 1.6272,
      "step": 1609
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7765272121436506e-05,
      "loss": 1.3997,
      "step": 1610
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7763791188448723e-05,
      "loss": 1.4054,
      "step": 1611
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.776231025546094e-05,
      "loss": 1.3849,
      "step": 1612
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7760829322473158e-05,
      "loss": 1.3395,
      "step": 1613
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.775934838948538e-05,
      "loss": 1.4332,
      "step": 1614
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7757867456497597e-05,
      "loss": 1.7074,
      "step": 1615
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.775638652350981e-05,
      "loss": 1.6395,
      "step": 1616
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.775490559052203e-05,
      "loss": 1.5595,
      "step": 1617
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.775342465753425e-05,
      "loss": 1.5073,
      "step": 1618
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7751943724546467e-05,
      "loss": 1.6767,
      "step": 1619
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.775046279155868e-05,
      "loss": 1.7866,
      "step": 1620
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7748981858570902e-05,
      "loss": 1.4641,
      "step": 1621
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.774750092558312e-05,
      "loss": 1.2825,
      "step": 1622
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7746019992595337e-05,
      "loss": 1.6672,
      "step": 1623
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7744539059607554e-05,
      "loss": 1.4687,
      "step": 1624
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7743058126619772e-05,
      "loss": 1.4554,
      "step": 1625
    },
    {
      "epoch": 0.36,
      "eval_loss": NaN,
      "eval_runtime": 4.3675,
      "eval_samples_per_second": 353.517,
      "eval_steps_per_second": 44.19,
      "step": 1625
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.774157719363199e-05,
      "loss": 1.4568,
      "step": 1626
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7740096260644207e-05,
      "loss": 1.6748,
      "step": 1627
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7738615327656424e-05,
      "loss": 1.4512,
      "step": 1628
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7737134394668642e-05,
      "loss": 1.4898,
      "step": 1629
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.773565346168086e-05,
      "loss": 1.6185,
      "step": 1630
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.773417252869308e-05,
      "loss": 1.4866,
      "step": 1631
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7732691595705294e-05,
      "loss": 1.3555,
      "step": 1632
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7731210662717512e-05,
      "loss": 1.5574,
      "step": 1633
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7729729729729733e-05,
      "loss": 1.6009,
      "step": 1634
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.772824879674195e-05,
      "loss": 1.6291,
      "step": 1635
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7726767863754168e-05,
      "loss": 1.6131,
      "step": 1636
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7725286930766382e-05,
      "loss": 1.3739,
      "step": 1637
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7723805997778603e-05,
      "loss": 1.573,
      "step": 1638
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.772232506479082e-05,
      "loss": 1.4915,
      "step": 1639
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7720844131803038e-05,
      "loss": 1.5739,
      "step": 1640
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7719363198815255e-05,
      "loss": 1.6276,
      "step": 1641
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7717882265827473e-05,
      "loss": 1.2811,
      "step": 1642
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.771640133283969e-05,
      "loss": 1.4072,
      "step": 1643
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7714920399851908e-05,
      "loss": 1.2039,
      "step": 1644
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7713439466864126e-05,
      "loss": 1.698,
      "step": 1645
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7711958533876343e-05,
      "loss": 1.4328,
      "step": 1646
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.771047760088856e-05,
      "loss": 1.4203,
      "step": 1647
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7708996667900778e-05,
      "loss": 1.6676,
      "step": 1648
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7707515734912996e-05,
      "loss": 1.6936,
      "step": 1649
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7706034801925213e-05,
      "loss": 1.5524,
      "step": 1650
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7704553868937434e-05,
      "loss": 1.7059,
      "step": 1651
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.770307293594965e-05,
      "loss": 1.4418,
      "step": 1652
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7701592002961866e-05,
      "loss": 1.3797,
      "step": 1653
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7700111069974083e-05,
      "loss": 1.3289,
      "step": 1654
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.7698630136986304e-05,
      "loss": 1.4162,
      "step": 1655
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.769714920399852e-05,
      "loss": 1.4224,
      "step": 1656
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.769566827101074e-05,
      "loss": 1.639,
      "step": 1657
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7694187338022957e-05,
      "loss": 1.4095,
      "step": 1658
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7692706405035174e-05,
      "loss": 1.3955,
      "step": 1659
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.769122547204739e-05,
      "loss": 1.3081,
      "step": 1660
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.768974453905961e-05,
      "loss": 1.4568,
      "step": 1661
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7688263606071827e-05,
      "loss": 1.1886,
      "step": 1662
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7686782673084044e-05,
      "loss": 1.5486,
      "step": 1663
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7685301740096262e-05,
      "loss": 1.3035,
      "step": 1664
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.768382080710848e-05,
      "loss": 1.5688,
      "step": 1665
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7682339874120697e-05,
      "loss": 1.6887,
      "step": 1666
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7680858941132914e-05,
      "loss": 1.4102,
      "step": 1667
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7679378008145135e-05,
      "loss": 1.6843,
      "step": 1668
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.767789707515735e-05,
      "loss": 1.7495,
      "step": 1669
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7676416142169567e-05,
      "loss": 1.5654,
      "step": 1670
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7674935209181784e-05,
      "loss": 1.6141,
      "step": 1671
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7673454276194005e-05,
      "loss": 1.4828,
      "step": 1672
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7671973343206223e-05,
      "loss": 1.4924,
      "step": 1673
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7670492410218437e-05,
      "loss": 1.6497,
      "step": 1674
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7669011477230658e-05,
      "loss": 1.6114,
      "step": 1675
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7667530544242875e-05,
      "loss": 1.3515,
      "step": 1676
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7666049611255093e-05,
      "loss": 1.6859,
      "step": 1677
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.766456867826731e-05,
      "loss": 1.3168,
      "step": 1678
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7663087745279528e-05,
      "loss": 1.3895,
      "step": 1679
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7661606812291745e-05,
      "loss": 1.7946,
      "step": 1680
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7660125879303963e-05,
      "loss": 1.379,
      "step": 1681
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.765864494631618e-05,
      "loss": 1.5624,
      "step": 1682
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7657164013328398e-05,
      "loss": 1.888,
      "step": 1683
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7655683080340615e-05,
      "loss": 1.4912,
      "step": 1684
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7654202147352833e-05,
      "loss": 1.6269,
      "step": 1685
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.765272121436505e-05,
      "loss": 1.4182,
      "step": 1686
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7651240281377268e-05,
      "loss": 1.5744,
      "step": 1687
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7649759348389486e-05,
      "loss": 1.4767,
      "step": 1688
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7648278415401706e-05,
      "loss": 1.5683,
      "step": 1689
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.764679748241392e-05,
      "loss": 1.3483,
      "step": 1690
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7645316549426138e-05,
      "loss": 1.7441,
      "step": 1691
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.764383561643836e-05,
      "loss": 1.7307,
      "step": 1692
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7642354683450577e-05,
      "loss": 1.3136,
      "step": 1693
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7640873750462794e-05,
      "loss": 1.7392,
      "step": 1694
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.763939281747501e-05,
      "loss": 1.5951,
      "step": 1695
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.763791188448723e-05,
      "loss": 1.5817,
      "step": 1696
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7636430951499447e-05,
      "loss": 1.4966,
      "step": 1697
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.7634950018511664e-05,
      "loss": 1.445,
      "step": 1698
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.763346908552388e-05,
      "loss": 1.5977,
      "step": 1699
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.76319881525361e-05,
      "loss": 1.7746,
      "step": 1700
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7630507219548317e-05,
      "loss": 1.5373,
      "step": 1701
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7629026286560534e-05,
      "loss": 1.7498,
      "step": 1702
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7627545353572752e-05,
      "loss": 1.7311,
      "step": 1703
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.762606442058497e-05,
      "loss": 1.4181,
      "step": 1704
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7624583487597187e-05,
      "loss": 1.547,
      "step": 1705
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7623102554609404e-05,
      "loss": 1.6259,
      "step": 1706
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7621621621621622e-05,
      "loss": 1.4912,
      "step": 1707
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.762014068863384e-05,
      "loss": 1.5381,
      "step": 1708
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.761865975564606e-05,
      "loss": 1.4909,
      "step": 1709
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7617178822658278e-05,
      "loss": 1.3951,
      "step": 1710
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7615697889670492e-05,
      "loss": 1.5137,
      "step": 1711
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7614216956682713e-05,
      "loss": 1.3831,
      "step": 1712
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.761273602369493e-05,
      "loss": 1.4293,
      "step": 1713
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7611255090707148e-05,
      "loss": 1.5285,
      "step": 1714
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7609774157719365e-05,
      "loss": 1.4632,
      "step": 1715
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7608293224731583e-05,
      "loss": 1.3451,
      "step": 1716
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.76068122917438e-05,
      "loss": 1.4893,
      "step": 1717
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7605331358756018e-05,
      "loss": 1.4298,
      "step": 1718
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7603850425768235e-05,
      "loss": 1.5859,
      "step": 1719
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7602369492780453e-05,
      "loss": 1.3497,
      "step": 1720
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.760088855979267e-05,
      "loss": 1.6179,
      "step": 1721
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7599407626804888e-05,
      "loss": 1.5052,
      "step": 1722
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7597926693817105e-05,
      "loss": 1.5256,
      "step": 1723
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7596445760829323e-05,
      "loss": 1.4302,
      "step": 1724
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.759496482784154e-05,
      "loss": 1.6077,
      "step": 1725
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.759348389485376e-05,
      "loss": 1.5455,
      "step": 1726
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7592002961865975e-05,
      "loss": 1.3474,
      "step": 1727
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7590522028878193e-05,
      "loss": 1.4963,
      "step": 1728
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7589041095890414e-05,
      "loss": 1.4318,
      "step": 1729
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.758756016290263e-05,
      "loss": 1.4843,
      "step": 1730
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.758607922991485e-05,
      "loss": 1.2812,
      "step": 1731
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7584598296927063e-05,
      "loss": 1.6203,
      "step": 1732
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7583117363939284e-05,
      "loss": 1.512,
      "step": 1733
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.75816364309515e-05,
      "loss": 1.3887,
      "step": 1734
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.758015549796372e-05,
      "loss": 1.5049,
      "step": 1735
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7578674564975937e-05,
      "loss": 1.518,
      "step": 1736
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7577193631988154e-05,
      "loss": 1.5183,
      "step": 1737
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.757571269900037e-05,
      "loss": 1.4071,
      "step": 1738
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.757423176601259e-05,
      "loss": 1.4155,
      "step": 1739
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7572750833024807e-05,
      "loss": 1.5151,
      "step": 1740
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7571269900037024e-05,
      "loss": 1.5674,
      "step": 1741
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.756978896704924e-05,
      "loss": 1.5622,
      "step": 1742
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7568308034061463e-05,
      "loss": 1.4487,
      "step": 1743
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7566827101073677e-05,
      "loss": 1.6382,
      "step": 1744
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.7565346168085894e-05,
      "loss": 1.5948,
      "step": 1745
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7563865235098115e-05,
      "loss": 1.3791,
      "step": 1746
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7562384302110333e-05,
      "loss": 1.597,
      "step": 1747
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7560903369122547e-05,
      "loss": 1.4064,
      "step": 1748
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7559422436134764e-05,
      "loss": 1.5368,
      "step": 1749
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7557941503146985e-05,
      "loss": 1.3694,
      "step": 1750
    },
    {
      "epoch": 0.39,
      "eval_loss": NaN,
      "eval_runtime": 4.3804,
      "eval_samples_per_second": 352.479,
      "eval_steps_per_second": 44.06,
      "step": 1750
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7556460570159203e-05,
      "loss": 1.5313,
      "step": 1751
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.755497963717142e-05,
      "loss": 1.5388,
      "step": 1752
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7553498704183638e-05,
      "loss": 1.3247,
      "step": 1753
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7552017771195855e-05,
      "loss": 1.202,
      "step": 1754
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7550536838208073e-05,
      "loss": 1.5,
      "step": 1755
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.754905590522029e-05,
      "loss": 1.2853,
      "step": 1756
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7547574972232508e-05,
      "loss": 1.5212,
      "step": 1757
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7546094039244725e-05,
      "loss": 1.5969,
      "step": 1758
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7544613106256943e-05,
      "loss": 1.5819,
      "step": 1759
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.754313217326916e-05,
      "loss": 1.6451,
      "step": 1760
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7541651240281378e-05,
      "loss": 1.4307,
      "step": 1761
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7540170307293595e-05,
      "loss": 1.5136,
      "step": 1762
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7538689374305816e-05,
      "loss": 1.5083,
      "step": 1763
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7537208441318034e-05,
      "loss": 1.5181,
      "step": 1764
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7535727508330248e-05,
      "loss": 1.3093,
      "step": 1765
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7534246575342465e-05,
      "loss": 1.5276,
      "step": 1766
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7532765642354686e-05,
      "loss": 1.5336,
      "step": 1767
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7531284709366904e-05,
      "loss": 1.6062,
      "step": 1768
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7529803776379118e-05,
      "loss": 1.3004,
      "step": 1769
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.752832284339134e-05,
      "loss": 1.3598,
      "step": 1770
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7526841910403556e-05,
      "loss": 1.4017,
      "step": 1771
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7525360977415774e-05,
      "loss": 1.6487,
      "step": 1772
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.752388004442799e-05,
      "loss": 1.3662,
      "step": 1773
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.752239911144021e-05,
      "loss": 1.2981,
      "step": 1774
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7520918178452426e-05,
      "loss": 1.5058,
      "step": 1775
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7519437245464644e-05,
      "loss": 1.5274,
      "step": 1776
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.751795631247686e-05,
      "loss": 1.3551,
      "step": 1777
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.751647537948908e-05,
      "loss": 1.4464,
      "step": 1778
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7514994446501297e-05,
      "loss": 1.4573,
      "step": 1779
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7513513513513517e-05,
      "loss": 1.397,
      "step": 1780
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.751203258052573e-05,
      "loss": 1.4711,
      "step": 1781
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.751055164753795e-05,
      "loss": 1.3525,
      "step": 1782
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7509070714550167e-05,
      "loss": 1.5985,
      "step": 1783
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7507589781562388e-05,
      "loss": 1.7767,
      "step": 1784
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7506108848574605e-05,
      "loss": 1.656,
      "step": 1785
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.750462791558682e-05,
      "loss": 1.8904,
      "step": 1786
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.750314698259904e-05,
      "loss": 1.5197,
      "step": 1787
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7501666049611258e-05,
      "loss": 1.3088,
      "step": 1788
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7500185116623475e-05,
      "loss": 1.3585,
      "step": 1789
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.7498704183635693e-05,
      "loss": 1.7185,
      "step": 1790
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.749722325064791e-05,
      "loss": 1.3729,
      "step": 1791
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7495742317660128e-05,
      "loss": 1.3491,
      "step": 1792
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7494261384672345e-05,
      "loss": 1.7346,
      "step": 1793
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7492780451684563e-05,
      "loss": 1.3845,
      "step": 1794
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.749129951869678e-05,
      "loss": 1.5303,
      "step": 1795
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7489818585708998e-05,
      "loss": 1.6754,
      "step": 1796
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7488337652721215e-05,
      "loss": 1.5967,
      "step": 1797
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7486856719733433e-05,
      "loss": 1.5184,
      "step": 1798
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.748537578674565e-05,
      "loss": 1.6577,
      "step": 1799
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7483894853757868e-05,
      "loss": 1.591,
      "step": 1800
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.748241392077009e-05,
      "loss": 1.2992,
      "step": 1801
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7480932987782303e-05,
      "loss": 1.6575,
      "step": 1802
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.747945205479452e-05,
      "loss": 1.492,
      "step": 1803
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.747797112180674e-05,
      "loss": 1.4663,
      "step": 1804
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.747649018881896e-05,
      "loss": 1.4084,
      "step": 1805
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7475009255831176e-05,
      "loss": 1.7992,
      "step": 1806
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7473528322843394e-05,
      "loss": 1.4297,
      "step": 1807
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.747204738985561e-05,
      "loss": 1.5268,
      "step": 1808
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.747056645686783e-05,
      "loss": 1.261,
      "step": 1809
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7469085523880046e-05,
      "loss": 1.538,
      "step": 1810
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7467604590892264e-05,
      "loss": 1.5401,
      "step": 1811
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.746612365790448e-05,
      "loss": 1.5036,
      "step": 1812
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.74646427249167e-05,
      "loss": 1.5708,
      "step": 1813
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7463161791928916e-05,
      "loss": 1.3572,
      "step": 1814
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7461680858941134e-05,
      "loss": 1.3824,
      "step": 1815
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.746019992595335e-05,
      "loss": 1.6981,
      "step": 1816
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.745871899296557e-05,
      "loss": 1.2791,
      "step": 1817
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7457238059977786e-05,
      "loss": 1.4893,
      "step": 1818
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7455757126990004e-05,
      "loss": 1.3609,
      "step": 1819
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.745427619400222e-05,
      "loss": 1.3363,
      "step": 1820
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7452795261014442e-05,
      "loss": 1.6318,
      "step": 1821
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.745131432802666e-05,
      "loss": 1.4039,
      "step": 1822
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7449833395038874e-05,
      "loss": 1.41,
      "step": 1823
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7448352462051095e-05,
      "loss": 1.5823,
      "step": 1824
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7446871529063312e-05,
      "loss": 1.4932,
      "step": 1825
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.744539059607553e-05,
      "loss": 1.4756,
      "step": 1826
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7443909663087744e-05,
      "loss": 1.6566,
      "step": 1827
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7442428730099965e-05,
      "loss": 1.4582,
      "step": 1828
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7440947797112183e-05,
      "loss": 1.5725,
      "step": 1829
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.74394668641244e-05,
      "loss": 1.3406,
      "step": 1830
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7437985931136618e-05,
      "loss": 1.4214,
      "step": 1831
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7436504998148835e-05,
      "loss": 1.7014,
      "step": 1832
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7435024065161053e-05,
      "loss": 1.6076,
      "step": 1833
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.743354313217327e-05,
      "loss": 1.3086,
      "step": 1834
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7432062199185488e-05,
      "loss": 1.6109,
      "step": 1835
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.7430581266197705e-05,
      "loss": 1.5739,
      "step": 1836
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7429100333209923e-05,
      "loss": 1.4547,
      "step": 1837
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7427619400222144e-05,
      "loss": 1.3818,
      "step": 1838
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7426138467234358e-05,
      "loss": 1.3968,
      "step": 1839
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7424657534246575e-05,
      "loss": 1.3519,
      "step": 1840
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7423176601258796e-05,
      "loss": 1.5227,
      "step": 1841
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7421695668271014e-05,
      "loss": 1.5047,
      "step": 1842
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.742021473528323e-05,
      "loss": 1.6652,
      "step": 1843
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7418733802295445e-05,
      "loss": 1.6505,
      "step": 1844
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7417252869307666e-05,
      "loss": 1.4917,
      "step": 1845
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7415771936319884e-05,
      "loss": 1.4559,
      "step": 1846
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.74142910033321e-05,
      "loss": 1.4832,
      "step": 1847
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.741281007034432e-05,
      "loss": 1.3282,
      "step": 1848
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7411329137356536e-05,
      "loss": 1.2027,
      "step": 1849
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7409848204368754e-05,
      "loss": 1.4484,
      "step": 1850
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.740836727138097e-05,
      "loss": 1.4199,
      "step": 1851
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.740688633839319e-05,
      "loss": 1.6027,
      "step": 1852
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7405405405405406e-05,
      "loss": 1.4399,
      "step": 1853
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7403924472417624e-05,
      "loss": 1.2273,
      "step": 1854
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.740244353942984e-05,
      "loss": 1.488,
      "step": 1855
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.740096260644206e-05,
      "loss": 1.5359,
      "step": 1856
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7399481673454276e-05,
      "loss": 1.6683,
      "step": 1857
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7398000740466497e-05,
      "loss": 1.5642,
      "step": 1858
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7396519807478715e-05,
      "loss": 1.6179,
      "step": 1859
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.739503887449093e-05,
      "loss": 1.6543,
      "step": 1860
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7393557941503146e-05,
      "loss": 1.4247,
      "step": 1861
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7392077008515367e-05,
      "loss": 1.6156,
      "step": 1862
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7390596075527585e-05,
      "loss": 1.3382,
      "step": 1863
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7389115142539802e-05,
      "loss": 1.4886,
      "step": 1864
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.738763420955202e-05,
      "loss": 1.4979,
      "step": 1865
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7386153276564237e-05,
      "loss": 1.568,
      "step": 1866
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7384672343576455e-05,
      "loss": 1.5148,
      "step": 1867
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7383191410588672e-05,
      "loss": 1.4041,
      "step": 1868
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.738171047760089e-05,
      "loss": 1.5403,
      "step": 1869
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7380229544613108e-05,
      "loss": 1.3622,
      "step": 1870
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7378748611625325e-05,
      "loss": 1.5664,
      "step": 1871
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7377267678637543e-05,
      "loss": 1.4727,
      "step": 1872
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.737578674564976e-05,
      "loss": 1.4997,
      "step": 1873
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7374305812661978e-05,
      "loss": 1.4412,
      "step": 1874
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.73728248796742e-05,
      "loss": 1.4759,
      "step": 1875
    },
    {
      "epoch": 0.41,
      "eval_loss": NaN,
      "eval_runtime": 4.3765,
      "eval_samples_per_second": 352.791,
      "eval_steps_per_second": 44.099,
      "step": 1875
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7371343946686413e-05,
      "loss": 1.637,
      "step": 1876
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.736986301369863e-05,
      "loss": 1.3783,
      "step": 1877
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7368382080710848e-05,
      "loss": 1.3437,
      "step": 1878
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.736690114772307e-05,
      "loss": 1.6127,
      "step": 1879
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.7365420214735286e-05,
      "loss": 1.4507,
      "step": 1880
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.73639392817475e-05,
      "loss": 1.5566,
      "step": 1881
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.736245834875972e-05,
      "loss": 1.5405,
      "step": 1882
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.736097741577194e-05,
      "loss": 1.3702,
      "step": 1883
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7359496482784156e-05,
      "loss": 1.4119,
      "step": 1884
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7358015549796374e-05,
      "loss": 1.281,
      "step": 1885
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.735653461680859e-05,
      "loss": 1.43,
      "step": 1886
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.735505368382081e-05,
      "loss": 1.3853,
      "step": 1887
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7353572750833026e-05,
      "loss": 1.5828,
      "step": 1888
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7352091817845244e-05,
      "loss": 1.5047,
      "step": 1889
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.735061088485746e-05,
      "loss": 1.6135,
      "step": 1890
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.734912995186968e-05,
      "loss": 1.4347,
      "step": 1891
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.73476490188819e-05,
      "loss": 1.5703,
      "step": 1892
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7346168085894114e-05,
      "loss": 1.7938,
      "step": 1893
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.734468715290633e-05,
      "loss": 1.2648,
      "step": 1894
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.734320621991855e-05,
      "loss": 1.4554,
      "step": 1895
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.734172528693077e-05,
      "loss": 1.3863,
      "step": 1896
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7340244353942984e-05,
      "loss": 1.4427,
      "step": 1897
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.73387634209552e-05,
      "loss": 1.3658,
      "step": 1898
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7337282487967422e-05,
      "loss": 1.7027,
      "step": 1899
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.733580155497964e-05,
      "loss": 1.3293,
      "step": 1900
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7334320621991857e-05,
      "loss": 1.4738,
      "step": 1901
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.733283968900407e-05,
      "loss": 1.5159,
      "step": 1902
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7331358756016292e-05,
      "loss": 1.4289,
      "step": 1903
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.732987782302851e-05,
      "loss": 1.5667,
      "step": 1904
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7328396890040727e-05,
      "loss": 1.4389,
      "step": 1905
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7326915957052945e-05,
      "loss": 1.3932,
      "step": 1906
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7325435024065162e-05,
      "loss": 1.3704,
      "step": 1907
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.732395409107738e-05,
      "loss": 1.286,
      "step": 1908
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7322473158089597e-05,
      "loss": 1.5192,
      "step": 1909
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7320992225101815e-05,
      "loss": 1.1625,
      "step": 1910
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7319511292114032e-05,
      "loss": 1.3081,
      "step": 1911
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.731803035912625e-05,
      "loss": 1.5354,
      "step": 1912
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.731654942613847e-05,
      "loss": 1.5862,
      "step": 1913
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7315068493150685e-05,
      "loss": 1.3823,
      "step": 1914
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7313587560162903e-05,
      "loss": 1.7366,
      "step": 1915
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7312106627175123e-05,
      "loss": 1.504,
      "step": 1916
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.731062569418734e-05,
      "loss": 1.454,
      "step": 1917
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7309144761199555e-05,
      "loss": 1.3733,
      "step": 1918
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7307663828211776e-05,
      "loss": 1.5008,
      "step": 1919
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7306182895223994e-05,
      "loss": 1.6099,
      "step": 1920
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.730470196223621e-05,
      "loss": 1.5551,
      "step": 1921
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.730322102924843e-05,
      "loss": 1.4568,
      "step": 1922
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7301740096260646e-05,
      "loss": 1.3575,
      "step": 1923
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7300259163272864e-05,
      "loss": 1.374,
      "step": 1924
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.729877823028508e-05,
      "loss": 1.4349,
      "step": 1925
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.72972972972973e-05,
      "loss": 1.7656,
      "step": 1926
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.7295816364309516e-05,
      "loss": 1.6731,
      "step": 1927
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7294335431321734e-05,
      "loss": 1.4812,
      "step": 1928
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.729285449833395e-05,
      "loss": 1.5175,
      "step": 1929
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.729137356534617e-05,
      "loss": 1.4332,
      "step": 1930
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7289892632358386e-05,
      "loss": 1.4064,
      "step": 1931
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7288411699370604e-05,
      "loss": 1.6024,
      "step": 1932
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7286930766382825e-05,
      "loss": 1.2347,
      "step": 1933
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7285449833395042e-05,
      "loss": 1.4799,
      "step": 1934
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7283968900407256e-05,
      "loss": 1.3455,
      "step": 1935
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7282487967419477e-05,
      "loss": 1.4215,
      "step": 1936
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7281007034431695e-05,
      "loss": 1.3143,
      "step": 1937
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7279526101443912e-05,
      "loss": 1.5836,
      "step": 1938
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7278045168456126e-05,
      "loss": 1.3205,
      "step": 1939
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7276564235468347e-05,
      "loss": 1.6364,
      "step": 1940
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7275083302480565e-05,
      "loss": 1.5064,
      "step": 1941
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7273602369492782e-05,
      "loss": 1.295,
      "step": 1942
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7272121436505e-05,
      "loss": 1.5045,
      "step": 1943
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7270640503517217e-05,
      "loss": 1.6315,
      "step": 1944
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7269159570529435e-05,
      "loss": 1.6808,
      "step": 1945
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7267678637541652e-05,
      "loss": 1.5172,
      "step": 1946
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.726619770455387e-05,
      "loss": 1.6244,
      "step": 1947
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7264716771566087e-05,
      "loss": 1.2765,
      "step": 1948
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7263235838578305e-05,
      "loss": 1.2091,
      "step": 1949
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7261754905590526e-05,
      "loss": 1.3266,
      "step": 1950
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.726027397260274e-05,
      "loss": 1.4624,
      "step": 1951
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7258793039614957e-05,
      "loss": 1.3975,
      "step": 1952
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.725731210662718e-05,
      "loss": 1.2889,
      "step": 1953
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7255831173639396e-05,
      "loss": 1.4301,
      "step": 1954
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.725435024065161e-05,
      "loss": 1.307,
      "step": 1955
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7252869307663828e-05,
      "loss": 1.6316,
      "step": 1956
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.725138837467605e-05,
      "loss": 1.4855,
      "step": 1957
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7249907441688266e-05,
      "loss": 1.8153,
      "step": 1958
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7248426508700483e-05,
      "loss": 1.1986,
      "step": 1959
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.72469455757127e-05,
      "loss": 1.5259,
      "step": 1960
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.724546464272492e-05,
      "loss": 1.7585,
      "step": 1961
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7243983709737136e-05,
      "loss": 1.586,
      "step": 1962
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7242502776749354e-05,
      "loss": 1.6024,
      "step": 1963
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.724102184376157e-05,
      "loss": 1.2947,
      "step": 1964
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.723954091077379e-05,
      "loss": 1.7026,
      "step": 1965
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7238059977786006e-05,
      "loss": 1.3623,
      "step": 1966
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7236579044798224e-05,
      "loss": 1.6345,
      "step": 1967
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.723509811181044e-05,
      "loss": 1.3788,
      "step": 1968
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.723361717882266e-05,
      "loss": 1.3035,
      "step": 1969
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.723213624583488e-05,
      "loss": 1.5299,
      "step": 1970
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.7230655312847097e-05,
      "loss": 1.22,
      "step": 1971
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.722917437985931e-05,
      "loss": 1.3083,
      "step": 1972
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.722769344687153e-05,
      "loss": 1.5262,
      "step": 1973
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.722621251388375e-05,
      "loss": 1.375,
      "step": 1974
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7224731580895967e-05,
      "loss": 1.6753,
      "step": 1975
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.722325064790818e-05,
      "loss": 1.4522,
      "step": 1976
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7221769714920402e-05,
      "loss": 1.5469,
      "step": 1977
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.722028878193262e-05,
      "loss": 1.3901,
      "step": 1978
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7218807848944837e-05,
      "loss": 1.2507,
      "step": 1979
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7217326915957055e-05,
      "loss": 1.3355,
      "step": 1980
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7215845982969272e-05,
      "loss": 1.2988,
      "step": 1981
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.721436504998149e-05,
      "loss": 1.5695,
      "step": 1982
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7212884116993707e-05,
      "loss": 1.4173,
      "step": 1983
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7211403184005925e-05,
      "loss": 1.7149,
      "step": 1984
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7209922251018142e-05,
      "loss": 1.475,
      "step": 1985
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.720844131803036e-05,
      "loss": 1.5796,
      "step": 1986
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.720696038504258e-05,
      "loss": 1.7195,
      "step": 1987
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7205479452054795e-05,
      "loss": 1.4992,
      "step": 1988
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7203998519067012e-05,
      "loss": 1.4342,
      "step": 1989
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.720251758607923e-05,
      "loss": 1.2699,
      "step": 1990
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.720103665309145e-05,
      "loss": 1.4176,
      "step": 1991
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7199555720103668e-05,
      "loss": 1.4767,
      "step": 1992
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7198074787115882e-05,
      "loss": 1.349,
      "step": 1993
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7196593854128103e-05,
      "loss": 1.4304,
      "step": 1994
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.719511292114032e-05,
      "loss": 1.3842,
      "step": 1995
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.719363198815254e-05,
      "loss": 1.5055,
      "step": 1996
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7192151055164752e-05,
      "loss": 1.5837,
      "step": 1997
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7190670122176973e-05,
      "loss": 1.6659,
      "step": 1998
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.718918918918919e-05,
      "loss": 1.652,
      "step": 1999
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.718770825620141e-05,
      "loss": 1.5559,
      "step": 2000
    },
    {
      "epoch": 0.44,
      "eval_loss": NaN,
      "eval_runtime": 4.3705,
      "eval_samples_per_second": 353.274,
      "eval_steps_per_second": 44.159,
      "step": 2000
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7186227323213626e-05,
      "loss": 1.4338,
      "step": 2001
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7184746390225843e-05,
      "loss": 1.7284,
      "step": 2002
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.718326545723806e-05,
      "loss": 1.4846,
      "step": 2003
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.718178452425028e-05,
      "loss": 1.6388,
      "step": 2004
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7180303591262496e-05,
      "loss": 1.3395,
      "step": 2005
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7178822658274714e-05,
      "loss": 1.4306,
      "step": 2006
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.717734172528693e-05,
      "loss": 1.6539,
      "step": 2007
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7175860792299152e-05,
      "loss": 1.6527,
      "step": 2008
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7174379859311366e-05,
      "loss": 1.533,
      "step": 2009
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7172898926323584e-05,
      "loss": 1.346,
      "step": 2010
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7171417993335804e-05,
      "loss": 1.5626,
      "step": 2011
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7169937060348022e-05,
      "loss": 1.4638,
      "step": 2012
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.716845612736024e-05,
      "loss": 1.4103,
      "step": 2013
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7166975194372454e-05,
      "loss": 1.4166,
      "step": 2014
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7165494261384675e-05,
      "loss": 1.4059,
      "step": 2015
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7164013328396892e-05,
      "loss": 1.5458,
      "step": 2016
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.716253239540911e-05,
      "loss": 1.2488,
      "step": 2017
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.7161051462421327e-05,
      "loss": 1.5359,
      "step": 2018
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7159570529433545e-05,
      "loss": 1.6236,
      "step": 2019
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7158089596445762e-05,
      "loss": 1.4715,
      "step": 2020
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.715660866345798e-05,
      "loss": 1.3158,
      "step": 2021
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7155127730470197e-05,
      "loss": 1.4471,
      "step": 2022
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7153646797482415e-05,
      "loss": 1.3131,
      "step": 2023
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7152165864494632e-05,
      "loss": 1.623,
      "step": 2024
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.715068493150685e-05,
      "loss": 1.3262,
      "step": 2025
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7149203998519067e-05,
      "loss": 1.5626,
      "step": 2026
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7147723065531285e-05,
      "loss": 1.379,
      "step": 2027
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7146242132543506e-05,
      "loss": 1.4602,
      "step": 2028
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7144761199555723e-05,
      "loss": 1.5484,
      "step": 2029
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7143280266567937e-05,
      "loss": 1.2341,
      "step": 2030
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7141799333580155e-05,
      "loss": 1.498,
      "step": 2031
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7140318400592376e-05,
      "loss": 1.4162,
      "step": 2032
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7138837467604593e-05,
      "loss": 1.5798,
      "step": 2033
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.713735653461681e-05,
      "loss": 1.5204,
      "step": 2034
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7135875601629028e-05,
      "loss": 1.705,
      "step": 2035
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7134394668641246e-05,
      "loss": 1.5929,
      "step": 2036
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7132913735653463e-05,
      "loss": 1.5916,
      "step": 2037
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.713143280266568e-05,
      "loss": 1.6782,
      "step": 2038
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.71299518696779e-05,
      "loss": 1.5224,
      "step": 2039
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7128470936690116e-05,
      "loss": 1.3154,
      "step": 2040
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7126990003702333e-05,
      "loss": 1.5751,
      "step": 2041
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.712550907071455e-05,
      "loss": 1.4855,
      "step": 2042
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.712402813772677e-05,
      "loss": 1.378,
      "step": 2043
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7122547204738986e-05,
      "loss": 1.5289,
      "step": 2044
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7121066271751207e-05,
      "loss": 1.4905,
      "step": 2045
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.711958533876342e-05,
      "loss": 1.4216,
      "step": 2046
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.711810440577564e-05,
      "loss": 1.3379,
      "step": 2047
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.711662347278786e-05,
      "loss": 1.4661,
      "step": 2048
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7115142539800077e-05,
      "loss": 1.4623,
      "step": 2049
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7113661606812294e-05,
      "loss": 1.379,
      "step": 2050
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.711218067382451e-05,
      "loss": 1.6293,
      "step": 2051
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.711069974083673e-05,
      "loss": 1.58,
      "step": 2052
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7109218807848947e-05,
      "loss": 1.5702,
      "step": 2053
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7107737874861165e-05,
      "loss": 1.3953,
      "step": 2054
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7106256941873382e-05,
      "loss": 1.4711,
      "step": 2055
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.71047760088856e-05,
      "loss": 1.3462,
      "step": 2056
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7103295075897817e-05,
      "loss": 1.3133,
      "step": 2057
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7101814142910035e-05,
      "loss": 1.3616,
      "step": 2058
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7100333209922252e-05,
      "loss": 1.4452,
      "step": 2059
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.709885227693447e-05,
      "loss": 1.3057,
      "step": 2060
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7097371343946687e-05,
      "loss": 1.4017,
      "step": 2061
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7095890410958908e-05,
      "loss": 1.4496,
      "step": 2062
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.7094409477971122e-05,
      "loss": 1.3649,
      "step": 2063
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.709292854498334e-05,
      "loss": 1.4768,
      "step": 2064
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.709144761199556e-05,
      "loss": 1.5548,
      "step": 2065
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7089966679007778e-05,
      "loss": 1.4803,
      "step": 2066
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7088485746019992e-05,
      "loss": 1.2209,
      "step": 2067
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.708700481303221e-05,
      "loss": 1.3842,
      "step": 2068
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.708552388004443e-05,
      "loss": 1.3982,
      "step": 2069
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7084042947056648e-05,
      "loss": 1.5274,
      "step": 2070
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7082562014068866e-05,
      "loss": 1.6001,
      "step": 2071
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7081081081081083e-05,
      "loss": 1.5563,
      "step": 2072
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.70796001480933e-05,
      "loss": 1.4818,
      "step": 2073
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7078119215105518e-05,
      "loss": 1.4868,
      "step": 2074
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7076638282117736e-05,
      "loss": 1.6953,
      "step": 2075
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7075157349129953e-05,
      "loss": 1.3468,
      "step": 2076
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.707367641614217e-05,
      "loss": 1.3279,
      "step": 2077
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7072195483154388e-05,
      "loss": 1.6346,
      "step": 2078
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7070714550166606e-05,
      "loss": 1.4922,
      "step": 2079
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7069233617178823e-05,
      "loss": 1.6155,
      "step": 2080
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.706775268419104e-05,
      "loss": 1.5362,
      "step": 2081
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7066271751203262e-05,
      "loss": 1.5779,
      "step": 2082
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7064790818215476e-05,
      "loss": 1.4909,
      "step": 2083
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7063309885227693e-05,
      "loss": 1.4566,
      "step": 2084
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.706182895223991e-05,
      "loss": 1.5496,
      "step": 2085
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7060348019252132e-05,
      "loss": 1.2952,
      "step": 2086
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.705886708626435e-05,
      "loss": 1.5328,
      "step": 2087
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7057386153276563e-05,
      "loss": 1.6852,
      "step": 2088
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7055905220288784e-05,
      "loss": 1.5983,
      "step": 2089
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7054424287301002e-05,
      "loss": 1.6292,
      "step": 2090
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.705294335431322e-05,
      "loss": 1.5371,
      "step": 2091
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7051462421325437e-05,
      "loss": 1.4084,
      "step": 2092
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7049981488337654e-05,
      "loss": 1.3749,
      "step": 2093
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7048500555349872e-05,
      "loss": 1.3213,
      "step": 2094
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.704701962236209e-05,
      "loss": 1.4595,
      "step": 2095
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7045538689374307e-05,
      "loss": 1.6959,
      "step": 2096
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7044057756386525e-05,
      "loss": 1.3094,
      "step": 2097
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7042576823398742e-05,
      "loss": 1.3666,
      "step": 2098
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7041095890410963e-05,
      "loss": 1.2093,
      "step": 2099
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7039614957423177e-05,
      "loss": 1.3705,
      "step": 2100
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7038134024435395e-05,
      "loss": 1.5327,
      "step": 2101
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7036653091447612e-05,
      "loss": 1.3582,
      "step": 2102
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7035172158459833e-05,
      "loss": 1.4636,
      "step": 2103
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7033691225472047e-05,
      "loss": 1.5908,
      "step": 2104
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7032210292484265e-05,
      "loss": 1.3484,
      "step": 2105
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7030729359496486e-05,
      "loss": 1.5382,
      "step": 2106
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.7029248426508703e-05,
      "loss": 1.6737,
      "step": 2107
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.702776749352092e-05,
      "loss": 1.3574,
      "step": 2108
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7026286560533135e-05,
      "loss": 1.7893,
      "step": 2109
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7024805627545356e-05,
      "loss": 1.4148,
      "step": 2110
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7023324694557573e-05,
      "loss": 1.4673,
      "step": 2111
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.702184376156979e-05,
      "loss": 1.3898,
      "step": 2112
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7020362828582008e-05,
      "loss": 1.5374,
      "step": 2113
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7018881895594226e-05,
      "loss": 1.5877,
      "step": 2114
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7017400962606443e-05,
      "loss": 1.6196,
      "step": 2115
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.701592002961866e-05,
      "loss": 1.4628,
      "step": 2116
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7014439096630878e-05,
      "loss": 1.4357,
      "step": 2117
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7012958163643096e-05,
      "loss": 1.3217,
      "step": 2118
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7011477230655313e-05,
      "loss": 1.3643,
      "step": 2119
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7009996297667534e-05,
      "loss": 1.4565,
      "step": 2120
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7008515364679748e-05,
      "loss": 1.5372,
      "step": 2121
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7007034431691966e-05,
      "loss": 1.5506,
      "step": 2122
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7005553498704187e-05,
      "loss": 1.5722,
      "step": 2123
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7004072565716404e-05,
      "loss": 1.3197,
      "step": 2124
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.700259163272862e-05,
      "loss": 1.3598,
      "step": 2125
    },
    {
      "epoch": 0.47,
      "eval_loss": NaN,
      "eval_runtime": 4.3694,
      "eval_samples_per_second": 353.363,
      "eval_steps_per_second": 44.17,
      "step": 2125
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.7001110699740836e-05,
      "loss": 1.5794,
      "step": 2126
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6999629766753057e-05,
      "loss": 1.5574,
      "step": 2127
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6998148833765274e-05,
      "loss": 1.5881,
      "step": 2128
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6996667900777492e-05,
      "loss": 1.4533,
      "step": 2129
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.699518696778971e-05,
      "loss": 1.7238,
      "step": 2130
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6993706034801927e-05,
      "loss": 1.641,
      "step": 2131
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6992225101814144e-05,
      "loss": 1.469,
      "step": 2132
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6990744168826362e-05,
      "loss": 1.358,
      "step": 2133
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.698926323583858e-05,
      "loss": 1.666,
      "step": 2134
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6987782302850797e-05,
      "loss": 1.267,
      "step": 2135
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6986301369863014e-05,
      "loss": 1.6122,
      "step": 2136
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6984820436875232e-05,
      "loss": 1.4462,
      "step": 2137
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.698333950388745e-05,
      "loss": 1.375,
      "step": 2138
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6981858570899667e-05,
      "loss": 1.6163,
      "step": 2139
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6980377637911888e-05,
      "loss": 1.4971,
      "step": 2140
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6978896704924105e-05,
      "loss": 1.4615,
      "step": 2141
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.697741577193632e-05,
      "loss": 1.222,
      "step": 2142
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6975934838948537e-05,
      "loss": 1.2612,
      "step": 2143
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6974453905960758e-05,
      "loss": 1.5539,
      "step": 2144
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6972972972972975e-05,
      "loss": 1.3771,
      "step": 2145
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.697149203998519e-05,
      "loss": 1.506,
      "step": 2146
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.697001110699741e-05,
      "loss": 1.5902,
      "step": 2147
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6968530174009628e-05,
      "loss": 1.3308,
      "step": 2148
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6967049241021846e-05,
      "loss": 1.6754,
      "step": 2149
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6965568308034063e-05,
      "loss": 1.4028,
      "step": 2150
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.696408737504628e-05,
      "loss": 1.4837,
      "step": 2151
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6962606442058498e-05,
      "loss": 1.6128,
      "step": 2152
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6961125509070716e-05,
      "loss": 1.2707,
      "step": 2153
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.6959644576082933e-05,
      "loss": 1.5789,
      "step": 2154
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.695816364309515e-05,
      "loss": 1.6519,
      "step": 2155
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6956682710107368e-05,
      "loss": 1.6261,
      "step": 2156
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.695520177711959e-05,
      "loss": 1.7943,
      "step": 2157
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6953720844131803e-05,
      "loss": 1.6785,
      "step": 2158
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.695223991114402e-05,
      "loss": 1.4456,
      "step": 2159
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6950758978156238e-05,
      "loss": 1.5635,
      "step": 2160
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.694927804516846e-05,
      "loss": 1.294,
      "step": 2161
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6947797112180677e-05,
      "loss": 1.5173,
      "step": 2162
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.694631617919289e-05,
      "loss": 1.4875,
      "step": 2163
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.694483524620511e-05,
      "loss": 1.33,
      "step": 2164
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.694335431321733e-05,
      "loss": 1.5032,
      "step": 2165
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6941873380229547e-05,
      "loss": 1.629,
      "step": 2166
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6940392447241764e-05,
      "loss": 1.7603,
      "step": 2167
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6938911514253982e-05,
      "loss": 1.4517,
      "step": 2168
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.69374305812662e-05,
      "loss": 1.62,
      "step": 2169
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6935949648278417e-05,
      "loss": 1.3961,
      "step": 2170
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6934468715290634e-05,
      "loss": 1.4244,
      "step": 2171
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6932987782302852e-05,
      "loss": 1.4508,
      "step": 2172
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.693150684931507e-05,
      "loss": 1.4877,
      "step": 2173
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6930025916327287e-05,
      "loss": 1.4607,
      "step": 2174
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6928544983339504e-05,
      "loss": 1.5602,
      "step": 2175
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6927064050351722e-05,
      "loss": 1.6971,
      "step": 2176
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6925583117363943e-05,
      "loss": 1.5372,
      "step": 2177
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.692410218437616e-05,
      "loss": 1.4444,
      "step": 2178
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6922621251388374e-05,
      "loss": 1.3645,
      "step": 2179
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6921140318400592e-05,
      "loss": 1.4248,
      "step": 2180
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6919659385412813e-05,
      "loss": 1.3863,
      "step": 2181
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.691817845242503e-05,
      "loss": 1.3804,
      "step": 2182
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6916697519437248e-05,
      "loss": 1.2271,
      "step": 2183
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6915216586449465e-05,
      "loss": 1.5421,
      "step": 2184
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6913735653461683e-05,
      "loss": 1.3047,
      "step": 2185
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.69122547204739e-05,
      "loss": 1.3166,
      "step": 2186
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6910773787486118e-05,
      "loss": 1.2846,
      "step": 2187
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6909292854498335e-05,
      "loss": 1.7136,
      "step": 2188
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6907811921510553e-05,
      "loss": 1.3763,
      "step": 2189
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.690633098852277e-05,
      "loss": 1.4428,
      "step": 2190
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6904850055534988e-05,
      "loss": 1.5756,
      "step": 2191
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6903369122547206e-05,
      "loss": 1.4363,
      "step": 2192
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6901888189559423e-05,
      "loss": 1.3759,
      "step": 2193
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6900407256571644e-05,
      "loss": 1.4728,
      "step": 2194
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6898926323583858e-05,
      "loss": 1.5657,
      "step": 2195
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6897445390596076e-05,
      "loss": 1.4793,
      "step": 2196
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6895964457608293e-05,
      "loss": 1.5238,
      "step": 2197
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.6894483524620514e-05,
      "loss": 1.4411,
      "step": 2198
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.689300259163273e-05,
      "loss": 1.4682,
      "step": 2199
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6891521658644946e-05,
      "loss": 1.4799,
      "step": 2200
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6890040725657167e-05,
      "loss": 1.717,
      "step": 2201
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6888559792669384e-05,
      "loss": 1.5056,
      "step": 2202
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.68870788596816e-05,
      "loss": 1.3653,
      "step": 2203
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.688559792669382e-05,
      "loss": 1.4859,
      "step": 2204
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6884116993706037e-05,
      "loss": 1.3846,
      "step": 2205
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6882636060718254e-05,
      "loss": 1.6527,
      "step": 2206
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.688115512773047e-05,
      "loss": 1.3147,
      "step": 2207
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.687967419474269e-05,
      "loss": 1.2732,
      "step": 2208
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6878193261754907e-05,
      "loss": 1.5288,
      "step": 2209
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6876712328767124e-05,
      "loss": 1.3438,
      "step": 2210
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6875231395779342e-05,
      "loss": 1.4969,
      "step": 2211
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.687375046279156e-05,
      "loss": 1.4499,
      "step": 2212
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6872269529803777e-05,
      "loss": 1.5597,
      "step": 2213
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6870788596815994e-05,
      "loss": 1.4986,
      "step": 2214
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6869307663828215e-05,
      "loss": 1.647,
      "step": 2215
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.686782673084043e-05,
      "loss": 1.7229,
      "step": 2216
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6866345797852647e-05,
      "loss": 1.2073,
      "step": 2217
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6864864864864868e-05,
      "loss": 1.5014,
      "step": 2218
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6863383931877085e-05,
      "loss": 1.2872,
      "step": 2219
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6861902998889303e-05,
      "loss": 1.4917,
      "step": 2220
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6860422065901517e-05,
      "loss": 1.475,
      "step": 2221
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6858941132913738e-05,
      "loss": 1.4182,
      "step": 2222
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6857460199925955e-05,
      "loss": 1.8156,
      "step": 2223
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6855979266938173e-05,
      "loss": 1.2306,
      "step": 2224
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.685449833395039e-05,
      "loss": 1.5267,
      "step": 2225
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6853017400962608e-05,
      "loss": 1.4557,
      "step": 2226
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6851536467974825e-05,
      "loss": 1.4531,
      "step": 2227
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6850055534987043e-05,
      "loss": 1.3878,
      "step": 2228
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.684857460199926e-05,
      "loss": 1.4685,
      "step": 2229
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6847093669011478e-05,
      "loss": 1.3095,
      "step": 2230
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6845612736023695e-05,
      "loss": 1.6,
      "step": 2231
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6844131803035913e-05,
      "loss": 1.3844,
      "step": 2232
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.684265087004813e-05,
      "loss": 1.335,
      "step": 2233
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6841169937060348e-05,
      "loss": 1.432,
      "step": 2234
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.683968900407257e-05,
      "loss": 1.406,
      "step": 2235
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6838208071084786e-05,
      "loss": 1.3175,
      "step": 2236
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6836727138097e-05,
      "loss": 1.5946,
      "step": 2237
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6835246205109218e-05,
      "loss": 1.4947,
      "step": 2238
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.683376527212144e-05,
      "loss": 1.3778,
      "step": 2239
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6832284339133657e-05,
      "loss": 1.6197,
      "step": 2240
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6830803406145874e-05,
      "loss": 1.4828,
      "step": 2241
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.682932247315809e-05,
      "loss": 1.3356,
      "step": 2242
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.682784154017031e-05,
      "loss": 1.4605,
      "step": 2243
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6826360607182527e-05,
      "loss": 1.3189,
      "step": 2244
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6824879674194744e-05,
      "loss": 1.6262,
      "step": 2245
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.682339874120696e-05,
      "loss": 1.3051,
      "step": 2246
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.682191780821918e-05,
      "loss": 1.3089,
      "step": 2247
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6820436875231397e-05,
      "loss": 1.3348,
      "step": 2248
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6818955942243614e-05,
      "loss": 1.6021,
      "step": 2249
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.681747500925583e-05,
      "loss": 1.535,
      "step": 2250
    },
    {
      "epoch": 0.5,
      "eval_loss": NaN,
      "eval_runtime": 4.3764,
      "eval_samples_per_second": 352.803,
      "eval_steps_per_second": 44.1,
      "step": 2250
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.681599407626805e-05,
      "loss": 1.4131,
      "step": 2251
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.681451314328027e-05,
      "loss": 1.4925,
      "step": 2252
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6813032210292484e-05,
      "loss": 1.3343,
      "step": 2253
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6811551277304702e-05,
      "loss": 1.2905,
      "step": 2254
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.681007034431692e-05,
      "loss": 1.3577,
      "step": 2255
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.680858941132914e-05,
      "loss": 1.5917,
      "step": 2256
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6807108478341358e-05,
      "loss": 1.7265,
      "step": 2257
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6805627545353572e-05,
      "loss": 1.5481,
      "step": 2258
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6804146612365793e-05,
      "loss": 1.4476,
      "step": 2259
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.680266567937801e-05,
      "loss": 1.5387,
      "step": 2260
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6801184746390228e-05,
      "loss": 1.4796,
      "step": 2261
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6799703813402445e-05,
      "loss": 1.3142,
      "step": 2262
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6798222880414663e-05,
      "loss": 1.4492,
      "step": 2263
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.679674194742688e-05,
      "loss": 1.6521,
      "step": 2264
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6795261014439098e-05,
      "loss": 1.5935,
      "step": 2265
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6793780081451315e-05,
      "loss": 1.4782,
      "step": 2266
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6792299148463533e-05,
      "loss": 1.5743,
      "step": 2267
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.679081821547575e-05,
      "loss": 1.4458,
      "step": 2268
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.678933728248797e-05,
      "loss": 1.381,
      "step": 2269
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6787856349500185e-05,
      "loss": 1.3344,
      "step": 2270
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6786375416512403e-05,
      "loss": 1.5084,
      "step": 2271
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.678489448352462e-05,
      "loss": 1.3328,
      "step": 2272
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.678341355053684e-05,
      "loss": 1.6316,
      "step": 2273
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6781932617549055e-05,
      "loss": 1.4916,
      "step": 2274
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6780451684561273e-05,
      "loss": 1.4229,
      "step": 2275
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6778970751573494e-05,
      "loss": 1.4648,
      "step": 2276
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.677748981858571e-05,
      "loss": 1.5732,
      "step": 2277
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.677600888559793e-05,
      "loss": 1.4946,
      "step": 2278
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6774527952610146e-05,
      "loss": 1.4887,
      "step": 2279
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6773047019622364e-05,
      "loss": 1.3609,
      "step": 2280
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.677156608663458e-05,
      "loss": 1.8798,
      "step": 2281
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.67700851536468e-05,
      "loss": 1.386,
      "step": 2282
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6768604220659017e-05,
      "loss": 1.8981,
      "step": 2283
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6767123287671234e-05,
      "loss": 1.33,
      "step": 2284
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.676564235468345e-05,
      "loss": 1.317,
      "step": 2285
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.676416142169567e-05,
      "loss": 1.2194,
      "step": 2286
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6762680488707887e-05,
      "loss": 1.3409,
      "step": 2287
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6761199555720104e-05,
      "loss": 1.1237,
      "step": 2288
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.675971862273232e-05,
      "loss": 1.6389,
      "step": 2289
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.6758237689744543e-05,
      "loss": 1.7758,
      "step": 2290
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6756756756756757e-05,
      "loss": 1.7794,
      "step": 2291
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6755275823768974e-05,
      "loss": 1.5498,
      "step": 2292
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6753794890781195e-05,
      "loss": 1.5363,
      "step": 2293
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6752313957793413e-05,
      "loss": 1.4262,
      "step": 2294
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6750833024805627e-05,
      "loss": 1.479,
      "step": 2295
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6749352091817848e-05,
      "loss": 1.5393,
      "step": 2296
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6747871158830065e-05,
      "loss": 1.6702,
      "step": 2297
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6746390225842283e-05,
      "loss": 1.592,
      "step": 2298
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.67449092928545e-05,
      "loss": 1.4139,
      "step": 2299
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6743428359866718e-05,
      "loss": 1.5718,
      "step": 2300
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6741947426878935e-05,
      "loss": 1.5188,
      "step": 2301
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6740466493891153e-05,
      "loss": 1.5527,
      "step": 2302
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.673898556090337e-05,
      "loss": 1.5165,
      "step": 2303
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6737504627915588e-05,
      "loss": 1.3499,
      "step": 2304
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6736023694927805e-05,
      "loss": 1.4454,
      "step": 2305
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6734542761940023e-05,
      "loss": 1.3905,
      "step": 2306
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.673306182895224e-05,
      "loss": 1.4952,
      "step": 2307
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6731580895964458e-05,
      "loss": 1.4463,
      "step": 2308
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6730099962976675e-05,
      "loss": 1.5486,
      "step": 2309
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6728619029988896e-05,
      "loss": 1.7985,
      "step": 2310
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6727138097001114e-05,
      "loss": 1.5335,
      "step": 2311
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6725657164013328e-05,
      "loss": 1.3442,
      "step": 2312
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.672417623102555e-05,
      "loss": 1.4157,
      "step": 2313
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6722695298037766e-05,
      "loss": 1.3286,
      "step": 2314
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6721214365049984e-05,
      "loss": 1.5766,
      "step": 2315
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6719733432062198e-05,
      "loss": 1.4373,
      "step": 2316
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.671825249907442e-05,
      "loss": 1.5451,
      "step": 2317
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6716771566086636e-05,
      "loss": 1.7163,
      "step": 2318
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6715290633098854e-05,
      "loss": 1.4161,
      "step": 2319
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.671380970011107e-05,
      "loss": 1.3893,
      "step": 2320
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.671232876712329e-05,
      "loss": 1.3909,
      "step": 2321
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6710847834135506e-05,
      "loss": 1.4407,
      "step": 2322
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6709366901147724e-05,
      "loss": 1.3595,
      "step": 2323
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.670788596815994e-05,
      "loss": 1.2597,
      "step": 2324
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.670640503517216e-05,
      "loss": 1.6345,
      "step": 2325
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6704924102184377e-05,
      "loss": 1.4311,
      "step": 2326
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6703443169196597e-05,
      "loss": 1.3447,
      "step": 2327
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.670196223620881e-05,
      "loss": 1.4276,
      "step": 2328
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.670048130322103e-05,
      "loss": 1.4688,
      "step": 2329
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.669900037023325e-05,
      "loss": 1.2683,
      "step": 2330
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6697519437245468e-05,
      "loss": 1.4675,
      "step": 2331
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6696038504257685e-05,
      "loss": 1.7136,
      "step": 2332
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.66945575712699e-05,
      "loss": 1.2705,
      "step": 2333
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.669307663828212e-05,
      "loss": 1.5188,
      "step": 2334
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.6691595705294338e-05,
      "loss": 1.527,
      "step": 2335
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6690114772306555e-05,
      "loss": 1.4399,
      "step": 2336
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6688633839318773e-05,
      "loss": 1.5652,
      "step": 2337
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.668715290633099e-05,
      "loss": 1.442,
      "step": 2338
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6685671973343208e-05,
      "loss": 1.4115,
      "step": 2339
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6684191040355425e-05,
      "loss": 1.5335,
      "step": 2340
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6682710107367643e-05,
      "loss": 1.348,
      "step": 2341
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.668122917437986e-05,
      "loss": 1.5944,
      "step": 2342
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6679748241392078e-05,
      "loss": 1.7374,
      "step": 2343
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6678267308404295e-05,
      "loss": 1.8262,
      "step": 2344
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6676786375416513e-05,
      "loss": 1.5127,
      "step": 2345
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.667530544242873e-05,
      "loss": 1.7186,
      "step": 2346
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.667382450944095e-05,
      "loss": 1.3539,
      "step": 2347
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.667234357645317e-05,
      "loss": 1.4304,
      "step": 2348
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6670862643465383e-05,
      "loss": 1.524,
      "step": 2349
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.66693817104776e-05,
      "loss": 1.6496,
      "step": 2350
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.666790077748982e-05,
      "loss": 1.4765,
      "step": 2351
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.666641984450204e-05,
      "loss": 1.5143,
      "step": 2352
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6664938911514256e-05,
      "loss": 1.5797,
      "step": 2353
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6663457978526474e-05,
      "loss": 1.5294,
      "step": 2354
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.666197704553869e-05,
      "loss": 1.4168,
      "step": 2355
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.666049611255091e-05,
      "loss": 1.539,
      "step": 2356
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6659015179563126e-05,
      "loss": 1.4041,
      "step": 2357
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6657534246575344e-05,
      "loss": 1.5266,
      "step": 2358
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.665605331358756e-05,
      "loss": 1.2758,
      "step": 2359
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.665457238059978e-05,
      "loss": 1.6549,
      "step": 2360
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6653091447611996e-05,
      "loss": 1.3001,
      "step": 2361
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6651610514624214e-05,
      "loss": 1.4024,
      "step": 2362
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.665012958163643e-05,
      "loss": 1.2957,
      "step": 2363
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6648648648648652e-05,
      "loss": 1.5157,
      "step": 2364
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6647167715660866e-05,
      "loss": 1.4505,
      "step": 2365
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6645686782673084e-05,
      "loss": 1.7281,
      "step": 2366
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.66442058496853e-05,
      "loss": 1.64,
      "step": 2367
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6642724916697522e-05,
      "loss": 1.4882,
      "step": 2368
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.664124398370974e-05,
      "loss": 1.6074,
      "step": 2369
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6639763050721954e-05,
      "loss": 1.3492,
      "step": 2370
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6638282117734175e-05,
      "loss": 1.5113,
      "step": 2371
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6636801184746392e-05,
      "loss": 1.5643,
      "step": 2372
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.663532025175861e-05,
      "loss": 1.7081,
      "step": 2373
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6633839318770828e-05,
      "loss": 1.3428,
      "step": 2374
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6632358385783045e-05,
      "loss": 1.3844,
      "step": 2375
    },
    {
      "epoch": 0.52,
      "eval_loss": NaN,
      "eval_runtime": 4.3847,
      "eval_samples_per_second": 352.131,
      "eval_steps_per_second": 44.016,
      "step": 2375
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6630877452795263e-05,
      "loss": 1.6031,
      "step": 2376
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.662939651980748e-05,
      "loss": 1.6686,
      "step": 2377
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6627915586819698e-05,
      "loss": 1.3299,
      "step": 2378
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6626434653831915e-05,
      "loss": 1.2069,
      "step": 2379
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.6624953720844133e-05,
      "loss": 1.7141,
      "step": 2380
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.662347278785635e-05,
      "loss": 1.5458,
      "step": 2381
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6621991854868568e-05,
      "loss": 1.4541,
      "step": 2382
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6620510921880785e-05,
      "loss": 1.538,
      "step": 2383
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6619029988893003e-05,
      "loss": 1.4861,
      "step": 2384
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6617549055905224e-05,
      "loss": 1.4903,
      "step": 2385
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6616068122917438e-05,
      "loss": 1.2789,
      "step": 2386
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6614587189929655e-05,
      "loss": 1.4645,
      "step": 2387
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6613106256941876e-05,
      "loss": 1.466,
      "step": 2388
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6611625323954094e-05,
      "loss": 1.7539,
      "step": 2389
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.661014439096631e-05,
      "loss": 1.658,
      "step": 2390
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.660866345797853e-05,
      "loss": 1.5326,
      "step": 2391
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6607182524990746e-05,
      "loss": 1.4383,
      "step": 2392
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6605701592002964e-05,
      "loss": 1.43,
      "step": 2393
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.660422065901518e-05,
      "loss": 1.385,
      "step": 2394
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.66027397260274e-05,
      "loss": 1.2245,
      "step": 2395
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6601258793039616e-05,
      "loss": 1.602,
      "step": 2396
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6599777860051834e-05,
      "loss": 1.5677,
      "step": 2397
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.659829692706405e-05,
      "loss": 1.3203,
      "step": 2398
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.659681599407627e-05,
      "loss": 1.4136,
      "step": 2399
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6595335061088486e-05,
      "loss": 1.4582,
      "step": 2400
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6593854128100704e-05,
      "loss": 1.4291,
      "step": 2401
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.659237319511292e-05,
      "loss": 1.0798,
      "step": 2402
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.659089226212514e-05,
      "loss": 1.543,
      "step": 2403
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6589411329137356e-05,
      "loss": 1.2538,
      "step": 2404
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6587930396149577e-05,
      "loss": 1.6374,
      "step": 2405
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6586449463161795e-05,
      "loss": 1.3977,
      "step": 2406
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.658496853017401e-05,
      "loss": 1.4535,
      "step": 2407
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.658348759718623e-05,
      "loss": 1.3281,
      "step": 2408
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6582006664198447e-05,
      "loss": 1.5759,
      "step": 2409
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6580525731210665e-05,
      "loss": 1.652,
      "step": 2410
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6579044798222882e-05,
      "loss": 1.383,
      "step": 2411
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.65775638652351e-05,
      "loss": 1.2673,
      "step": 2412
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6576082932247317e-05,
      "loss": 1.7255,
      "step": 2413
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6574601999259535e-05,
      "loss": 1.2963,
      "step": 2414
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6573121066271752e-05,
      "loss": 1.6077,
      "step": 2415
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.657164013328397e-05,
      "loss": 1.4616,
      "step": 2416
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6570159200296188e-05,
      "loss": 1.4331,
      "step": 2417
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6568678267308405e-05,
      "loss": 1.487,
      "step": 2418
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6567197334320623e-05,
      "loss": 1.2117,
      "step": 2419
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.656571640133284e-05,
      "loss": 1.4785,
      "step": 2420
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6564235468345058e-05,
      "loss": 1.4375,
      "step": 2421
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.656275453535728e-05,
      "loss": 1.3807,
      "step": 2422
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.6561273602369493e-05,
      "loss": 1.5554,
      "step": 2423
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.655979266938171e-05,
      "loss": 1.5036,
      "step": 2424
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.655831173639393e-05,
      "loss": 1.3454,
      "step": 2425
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.655683080340615e-05,
      "loss": 1.5206,
      "step": 2426
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6555349870418366e-05,
      "loss": 1.3943,
      "step": 2427
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.655386893743058e-05,
      "loss": 1.6983,
      "step": 2428
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.65523880044428e-05,
      "loss": 1.5821,
      "step": 2429
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.655090707145502e-05,
      "loss": 1.6515,
      "step": 2430
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6549426138467236e-05,
      "loss": 1.5475,
      "step": 2431
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6547945205479454e-05,
      "loss": 1.4632,
      "step": 2432
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.654646427249167e-05,
      "loss": 1.3008,
      "step": 2433
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.654498333950389e-05,
      "loss": 1.5998,
      "step": 2434
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6543502406516106e-05,
      "loss": 1.4072,
      "step": 2435
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6542021473528324e-05,
      "loss": 1.5875,
      "step": 2436
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.654054054054054e-05,
      "loss": 1.6999,
      "step": 2437
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.653905960755276e-05,
      "loss": 1.6688,
      "step": 2438
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.653757867456498e-05,
      "loss": 1.3712,
      "step": 2439
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6536097741577194e-05,
      "loss": 1.2505,
      "step": 2440
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.653461680858941e-05,
      "loss": 1.4553,
      "step": 2441
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6533135875601632e-05,
      "loss": 1.5939,
      "step": 2442
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.653165494261385e-05,
      "loss": 1.3115,
      "step": 2443
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6530174009626064e-05,
      "loss": 1.5374,
      "step": 2444
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.652869307663828e-05,
      "loss": 1.3944,
      "step": 2445
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6527212143650502e-05,
      "loss": 1.4377,
      "step": 2446
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.652573121066272e-05,
      "loss": 1.5471,
      "step": 2447
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6524250277674937e-05,
      "loss": 1.3895,
      "step": 2448
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6522769344687155e-05,
      "loss": 1.7352,
      "step": 2449
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6521288411699372e-05,
      "loss": 1.5428,
      "step": 2450
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.651980747871159e-05,
      "loss": 1.3178,
      "step": 2451
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6518326545723807e-05,
      "loss": 1.5024,
      "step": 2452
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6516845612736025e-05,
      "loss": 1.6058,
      "step": 2453
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6515364679748242e-05,
      "loss": 1.4985,
      "step": 2454
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.651388374676046e-05,
      "loss": 1.4934,
      "step": 2455
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6512402813772677e-05,
      "loss": 1.661,
      "step": 2456
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6510921880784895e-05,
      "loss": 1.4048,
      "step": 2457
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6509440947797112e-05,
      "loss": 1.3832,
      "step": 2458
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6507960014809333e-05,
      "loss": 1.5304,
      "step": 2459
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.650647908182155e-05,
      "loss": 1.3005,
      "step": 2460
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6504998148833765e-05,
      "loss": 1.4525,
      "step": 2461
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6503517215845983e-05,
      "loss": 1.4671,
      "step": 2462
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6502036282858203e-05,
      "loss": 1.4845,
      "step": 2463
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.650055534987042e-05,
      "loss": 1.563,
      "step": 2464
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6499074416882635e-05,
      "loss": 1.482,
      "step": 2465
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6497593483894856e-05,
      "loss": 1.3232,
      "step": 2466
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6496112550907074e-05,
      "loss": 1.4732,
      "step": 2467
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.649463161791929e-05,
      "loss": 1.4094,
      "step": 2468
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.649315068493151e-05,
      "loss": 1.4179,
      "step": 2469
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6491669751943726e-05,
      "loss": 1.2315,
      "step": 2470
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.6490188818955944e-05,
      "loss": 1.6456,
      "step": 2471
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.648870788596816e-05,
      "loss": 1.6596,
      "step": 2472
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.648722695298038e-05,
      "loss": 1.2624,
      "step": 2473
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6485746019992596e-05,
      "loss": 1.5356,
      "step": 2474
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6484265087004814e-05,
      "loss": 1.5602,
      "step": 2475
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6482784154017035e-05,
      "loss": 1.5355,
      "step": 2476
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.648130322102925e-05,
      "loss": 1.2531,
      "step": 2477
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6479822288041466e-05,
      "loss": 1.5241,
      "step": 2478
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6478341355053684e-05,
      "loss": 1.7319,
      "step": 2479
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6476860422065905e-05,
      "loss": 1.4902,
      "step": 2480
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6475379489078122e-05,
      "loss": 1.4981,
      "step": 2481
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6473898556090336e-05,
      "loss": 1.6462,
      "step": 2482
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6472417623102557e-05,
      "loss": 1.5801,
      "step": 2483
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6470936690114775e-05,
      "loss": 1.691,
      "step": 2484
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6469455757126992e-05,
      "loss": 1.7008,
      "step": 2485
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.646797482413921e-05,
      "loss": 1.4206,
      "step": 2486
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6466493891151427e-05,
      "loss": 1.4334,
      "step": 2487
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6465012958163645e-05,
      "loss": 1.5819,
      "step": 2488
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6463532025175862e-05,
      "loss": 1.4813,
      "step": 2489
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.646205109218808e-05,
      "loss": 1.4993,
      "step": 2490
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6460570159200297e-05,
      "loss": 1.5177,
      "step": 2491
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6459089226212515e-05,
      "loss": 1.1995,
      "step": 2492
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6457608293224732e-05,
      "loss": 1.5778,
      "step": 2493
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.645612736023695e-05,
      "loss": 1.4573,
      "step": 2494
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6454646427249167e-05,
      "loss": 1.59,
      "step": 2495
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6453165494261385e-05,
      "loss": 1.3432,
      "step": 2496
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6451684561273606e-05,
      "loss": 1.6385,
      "step": 2497
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.645020362828582e-05,
      "loss": 1.3718,
      "step": 2498
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6448722695298037e-05,
      "loss": 1.3297,
      "step": 2499
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.644724176231026e-05,
      "loss": 1.6352,
      "step": 2500
    },
    {
      "epoch": 0.55,
      "eval_loss": NaN,
      "eval_runtime": 4.3762,
      "eval_samples_per_second": 352.815,
      "eval_steps_per_second": 44.102,
      "step": 2500
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6445760829322476e-05,
      "loss": 1.4863,
      "step": 2501
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.644427989633469e-05,
      "loss": 1.3853,
      "step": 2502
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.644279896334691e-05,
      "loss": 1.4313,
      "step": 2503
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.644131803035913e-05,
      "loss": 1.4338,
      "step": 2504
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6439837097371346e-05,
      "loss": 1.3735,
      "step": 2505
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6438356164383563e-05,
      "loss": 1.48,
      "step": 2506
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.643687523139578e-05,
      "loss": 1.4818,
      "step": 2507
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6435394298408e-05,
      "loss": 1.4644,
      "step": 2508
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6433913365420216e-05,
      "loss": 1.4277,
      "step": 2509
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6432432432432434e-05,
      "loss": 1.5636,
      "step": 2510
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.643095149944465e-05,
      "loss": 1.5023,
      "step": 2511
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.642947056645687e-05,
      "loss": 1.2224,
      "step": 2512
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6427989633469086e-05,
      "loss": 1.437,
      "step": 2513
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.6426508700481304e-05,
      "loss": 1.5013,
      "step": 2514
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.642502776749352e-05,
      "loss": 1.6324,
      "step": 2515
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.642354683450574e-05,
      "loss": 1.3156,
      "step": 2516
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.642206590151796e-05,
      "loss": 1.5565,
      "step": 2517
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6420584968530177e-05,
      "loss": 1.4855,
      "step": 2518
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.641910403554239e-05,
      "loss": 1.5848,
      "step": 2519
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6417623102554612e-05,
      "loss": 1.4212,
      "step": 2520
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.641614216956683e-05,
      "loss": 1.4564,
      "step": 2521
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6414661236579047e-05,
      "loss": 1.3242,
      "step": 2522
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.641318030359126e-05,
      "loss": 1.5495,
      "step": 2523
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6411699370603482e-05,
      "loss": 1.3981,
      "step": 2524
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.64102184376157e-05,
      "loss": 1.5673,
      "step": 2525
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6408737504627917e-05,
      "loss": 1.5291,
      "step": 2526
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6407256571640135e-05,
      "loss": 1.3499,
      "step": 2527
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6405775638652352e-05,
      "loss": 1.2876,
      "step": 2528
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.640429470566457e-05,
      "loss": 1.5344,
      "step": 2529
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6402813772676787e-05,
      "loss": 1.5206,
      "step": 2530
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6401332839689005e-05,
      "loss": 1.436,
      "step": 2531
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6399851906701222e-05,
      "loss": 1.344,
      "step": 2532
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.639837097371344e-05,
      "loss": 1.5746,
      "step": 2533
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.639689004072566e-05,
      "loss": 1.786,
      "step": 2534
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6395409107737875e-05,
      "loss": 1.3916,
      "step": 2535
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6393928174750092e-05,
      "loss": 1.3774,
      "step": 2536
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6392447241762313e-05,
      "loss": 1.5312,
      "step": 2537
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.639096630877453e-05,
      "loss": 1.5011,
      "step": 2538
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6389485375786748e-05,
      "loss": 1.3174,
      "step": 2539
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6388004442798962e-05,
      "loss": 1.4868,
      "step": 2540
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6386523509811183e-05,
      "loss": 1.4236,
      "step": 2541
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.63850425768234e-05,
      "loss": 1.3258,
      "step": 2542
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.638356164383562e-05,
      "loss": 1.6838,
      "step": 2543
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6382080710847836e-05,
      "loss": 1.2268,
      "step": 2544
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6380599777860053e-05,
      "loss": 1.4818,
      "step": 2545
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.637911884487227e-05,
      "loss": 1.6077,
      "step": 2546
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.637763791188449e-05,
      "loss": 1.5019,
      "step": 2547
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6376156978896706e-05,
      "loss": 1.3105,
      "step": 2548
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6374676045908923e-05,
      "loss": 1.2886,
      "step": 2549
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.637319511292114e-05,
      "loss": 1.3364,
      "step": 2550
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.637171417993336e-05,
      "loss": 1.4632,
      "step": 2551
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6370233246945576e-05,
      "loss": 1.5589,
      "step": 2552
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6368752313957794e-05,
      "loss": 1.4023,
      "step": 2553
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6367271380970014e-05,
      "loss": 1.571,
      "step": 2554
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6365790447982232e-05,
      "loss": 1.5223,
      "step": 2555
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6364309514994446e-05,
      "loss": 1.2947,
      "step": 2556
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6362828582006664e-05,
      "loss": 1.4266,
      "step": 2557
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6361347649018884e-05,
      "loss": 1.5401,
      "step": 2558
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6359866716031102e-05,
      "loss": 1.4141,
      "step": 2559
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.635838578304332e-05,
      "loss": 1.4321,
      "step": 2560
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6356904850055537e-05,
      "loss": 1.2087,
      "step": 2561
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.6355423917067755e-05,
      "loss": 1.5413,
      "step": 2562
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6353942984079972e-05,
      "loss": 1.568,
      "step": 2563
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.635246205109219e-05,
      "loss": 1.7694,
      "step": 2564
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6350981118104407e-05,
      "loss": 1.3819,
      "step": 2565
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6349500185116625e-05,
      "loss": 1.7406,
      "step": 2566
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6348019252128842e-05,
      "loss": 1.5609,
      "step": 2567
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.634653831914106e-05,
      "loss": 1.4582,
      "step": 2568
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6345057386153277e-05,
      "loss": 1.5717,
      "step": 2569
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6343576453165495e-05,
      "loss": 1.4026,
      "step": 2570
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6342095520177716e-05,
      "loss": 1.3858,
      "step": 2571
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.634061458718993e-05,
      "loss": 1.4773,
      "step": 2572
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6339133654202147e-05,
      "loss": 1.4893,
      "step": 2573
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6337652721214365e-05,
      "loss": 1.2431,
      "step": 2574
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6336171788226586e-05,
      "loss": 1.3522,
      "step": 2575
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6334690855238803e-05,
      "loss": 1.2824,
      "step": 2576
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6333209922251017e-05,
      "loss": 1.555,
      "step": 2577
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6331728989263238e-05,
      "loss": 1.3645,
      "step": 2578
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6330248056275456e-05,
      "loss": 1.4981,
      "step": 2579
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6328767123287673e-05,
      "loss": 1.3893,
      "step": 2580
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.632728619029989e-05,
      "loss": 1.656,
      "step": 2581
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6325805257312108e-05,
      "loss": 1.287,
      "step": 2582
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6324324324324326e-05,
      "loss": 1.3908,
      "step": 2583
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6322843391336543e-05,
      "loss": 1.3219,
      "step": 2584
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.632136245834876e-05,
      "loss": 1.4634,
      "step": 2585
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.631988152536098e-05,
      "loss": 1.3848,
      "step": 2586
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6318400592373196e-05,
      "loss": 1.5985,
      "step": 2587
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6316919659385417e-05,
      "loss": 1.3654,
      "step": 2588
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.631543872639763e-05,
      "loss": 1.5628,
      "step": 2589
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.631395779340985e-05,
      "loss": 1.732,
      "step": 2590
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6312476860422066e-05,
      "loss": 1.4329,
      "step": 2591
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6310995927434287e-05,
      "loss": 1.512,
      "step": 2592
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.63095149944465e-05,
      "loss": 1.4301,
      "step": 2593
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.630803406145872e-05,
      "loss": 1.1357,
      "step": 2594
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.630655312847094e-05,
      "loss": 1.4881,
      "step": 2595
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6305072195483157e-05,
      "loss": 1.5664,
      "step": 2596
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6303591262495374e-05,
      "loss": 1.5258,
      "step": 2597
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6302110329507592e-05,
      "loss": 1.4699,
      "step": 2598
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.630062939651981e-05,
      "loss": 1.5171,
      "step": 2599
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6299148463532027e-05,
      "loss": 1.6007,
      "step": 2600
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6297667530544244e-05,
      "loss": 1.4377,
      "step": 2601
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6296186597556462e-05,
      "loss": 1.5061,
      "step": 2602
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.629470566456868e-05,
      "loss": 1.2091,
      "step": 2603
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6293224731580897e-05,
      "loss": 1.3846,
      "step": 2604
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6291743798593115e-05,
      "loss": 1.461,
      "step": 2605
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.6290262865605332e-05,
      "loss": 1.421,
      "step": 2606
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.628878193261755e-05,
      "loss": 1.3652,
      "step": 2607
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6287300999629767e-05,
      "loss": 1.4577,
      "step": 2608
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6285820066641988e-05,
      "loss": 1.4473,
      "step": 2609
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6284339133654202e-05,
      "loss": 1.3718,
      "step": 2610
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.628285820066642e-05,
      "loss": 1.4723,
      "step": 2611
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.628137726767864e-05,
      "loss": 1.3855,
      "step": 2612
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6279896334690858e-05,
      "loss": 1.5857,
      "step": 2613
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6278415401703072e-05,
      "loss": 1.4045,
      "step": 2614
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6276934468715293e-05,
      "loss": 1.4667,
      "step": 2615
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.627545353572751e-05,
      "loss": 1.6521,
      "step": 2616
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6273972602739728e-05,
      "loss": 1.5693,
      "step": 2617
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6272491669751946e-05,
      "loss": 1.6652,
      "step": 2618
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6271010736764163e-05,
      "loss": 1.4019,
      "step": 2619
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.626952980377638e-05,
      "loss": 1.5241,
      "step": 2620
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6268048870788598e-05,
      "loss": 1.4417,
      "step": 2621
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6266567937800816e-05,
      "loss": 1.5485,
      "step": 2622
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6265087004813033e-05,
      "loss": 1.4704,
      "step": 2623
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.626360607182525e-05,
      "loss": 1.362,
      "step": 2624
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6262125138837468e-05,
      "loss": 1.3591,
      "step": 2625
    },
    {
      "epoch": 0.58,
      "eval_loss": NaN,
      "eval_runtime": 4.3782,
      "eval_samples_per_second": 352.656,
      "eval_steps_per_second": 44.082,
      "step": 2625
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6260644205849686e-05,
      "loss": 1.3404,
      "step": 2626
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6259163272861903e-05,
      "loss": 1.4215,
      "step": 2627
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.625768233987412e-05,
      "loss": 1.6258,
      "step": 2628
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6256201406886342e-05,
      "loss": 1.4465,
      "step": 2629
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6254720473898556e-05,
      "loss": 1.4111,
      "step": 2630
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6253239540910773e-05,
      "loss": 1.4067,
      "step": 2631
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6251758607922994e-05,
      "loss": 1.4061,
      "step": 2632
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6250277674935212e-05,
      "loss": 1.3127,
      "step": 2633
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.624879674194743e-05,
      "loss": 1.5603,
      "step": 2634
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6247315808959643e-05,
      "loss": 1.5237,
      "step": 2635
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6245834875971864e-05,
      "loss": 1.8453,
      "step": 2636
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6244353942984082e-05,
      "loss": 1.4258,
      "step": 2637
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.62428730099963e-05,
      "loss": 1.4696,
      "step": 2638
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6241392077008517e-05,
      "loss": 1.5329,
      "step": 2639
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6239911144020734e-05,
      "loss": 1.3312,
      "step": 2640
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6238430211032952e-05,
      "loss": 1.2168,
      "step": 2641
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.623694927804517e-05,
      "loss": 1.6938,
      "step": 2642
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6235468345057387e-05,
      "loss": 1.0984,
      "step": 2643
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6233987412069605e-05,
      "loss": 1.291,
      "step": 2644
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6232506479081822e-05,
      "loss": 1.2505,
      "step": 2645
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6231025546094043e-05,
      "loss": 1.4031,
      "step": 2646
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6229544613106257e-05,
      "loss": 1.516,
      "step": 2647
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6228063680118475e-05,
      "loss": 1.6101,
      "step": 2648
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6226582747130695e-05,
      "loss": 1.4149,
      "step": 2649
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6225101814142913e-05,
      "loss": 1.2972,
      "step": 2650
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6223620881155127e-05,
      "loss": 1.2962,
      "step": 2651
    },
    {
      "epoch": 0.58,
      "learning_rate": 1.6222139948167345e-05,
      "loss": 1.3818,
      "step": 2652
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6220659015179566e-05,
      "loss": 1.513,
      "step": 2653
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6219178082191783e-05,
      "loss": 1.5608,
      "step": 2654
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6217697149204e-05,
      "loss": 1.3584,
      "step": 2655
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6216216216216218e-05,
      "loss": 1.3824,
      "step": 2656
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6214735283228436e-05,
      "loss": 1.4124,
      "step": 2657
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6213254350240653e-05,
      "loss": 1.7151,
      "step": 2658
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.621177341725287e-05,
      "loss": 1.5299,
      "step": 2659
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6210292484265088e-05,
      "loss": 1.5035,
      "step": 2660
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6208811551277306e-05,
      "loss": 1.4577,
      "step": 2661
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6207330618289523e-05,
      "loss": 1.4786,
      "step": 2662
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.620584968530174e-05,
      "loss": 1.3369,
      "step": 2663
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6204368752313958e-05,
      "loss": 1.3142,
      "step": 2664
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6202887819326176e-05,
      "loss": 1.4605,
      "step": 2665
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6201406886338397e-05,
      "loss": 1.4208,
      "step": 2666
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6199925953350614e-05,
      "loss": 1.5656,
      "step": 2667
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6198445020362828e-05,
      "loss": 1.2027,
      "step": 2668
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6196964087375046e-05,
      "loss": 1.7403,
      "step": 2669
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6195483154387267e-05,
      "loss": 1.5053,
      "step": 2670
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6194002221399484e-05,
      "loss": 1.4943,
      "step": 2671
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.61925212884117e-05,
      "loss": 1.369,
      "step": 2672
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.619104035542392e-05,
      "loss": 1.6131,
      "step": 2673
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6189559422436137e-05,
      "loss": 1.6987,
      "step": 2674
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6188078489448354e-05,
      "loss": 1.5251,
      "step": 2675
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6186597556460572e-05,
      "loss": 1.9183,
      "step": 2676
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.618511662347279e-05,
      "loss": 1.5549,
      "step": 2677
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6183635690485007e-05,
      "loss": 1.6805,
      "step": 2678
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6182154757497224e-05,
      "loss": 1.5208,
      "step": 2679
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6180673824509442e-05,
      "loss": 1.6801,
      "step": 2680
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.617919289152166e-05,
      "loss": 1.209,
      "step": 2681
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6177711958533877e-05,
      "loss": 1.5337,
      "step": 2682
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6176231025546098e-05,
      "loss": 1.6424,
      "step": 2683
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6174750092558312e-05,
      "loss": 1.4282,
      "step": 2684
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.617326915957053e-05,
      "loss": 1.5896,
      "step": 2685
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6171788226582747e-05,
      "loss": 1.4157,
      "step": 2686
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6170307293594968e-05,
      "loss": 1.3579,
      "step": 2687
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6168826360607185e-05,
      "loss": 1.3921,
      "step": 2688
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.61673454276194e-05,
      "loss": 1.4414,
      "step": 2689
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.616586449463162e-05,
      "loss": 1.4285,
      "step": 2690
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6164383561643838e-05,
      "loss": 1.272,
      "step": 2691
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6162902628656055e-05,
      "loss": 1.3119,
      "step": 2692
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.616142169566827e-05,
      "loss": 1.5045,
      "step": 2693
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.615994076268049e-05,
      "loss": 1.3876,
      "step": 2694
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6158459829692708e-05,
      "loss": 1.4162,
      "step": 2695
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6156978896704926e-05,
      "loss": 1.3745,
      "step": 2696
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.6155497963717143e-05,
      "loss": 1.2961,
      "step": 2697
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.615401703072936e-05,
      "loss": 1.4209,
      "step": 2698
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6152536097741578e-05,
      "loss": 1.4492,
      "step": 2699
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6151055164753796e-05,
      "loss": 1.5729,
      "step": 2700
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6149574231766013e-05,
      "loss": 1.1763,
      "step": 2701
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.614809329877823e-05,
      "loss": 1.3165,
      "step": 2702
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6146612365790448e-05,
      "loss": 1.4612,
      "step": 2703
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.614513143280267e-05,
      "loss": 1.5603,
      "step": 2704
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6143650499814883e-05,
      "loss": 1.3548,
      "step": 2705
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.61421695668271e-05,
      "loss": 1.3133,
      "step": 2706
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.614068863383932e-05,
      "loss": 1.206,
      "step": 2707
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.613920770085154e-05,
      "loss": 1.5193,
      "step": 2708
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6137726767863757e-05,
      "loss": 1.5226,
      "step": 2709
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6136245834875974e-05,
      "loss": 1.5901,
      "step": 2710
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.613476490188819e-05,
      "loss": 1.4196,
      "step": 2711
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.613328396890041e-05,
      "loss": 1.4021,
      "step": 2712
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6131803035912627e-05,
      "loss": 1.2019,
      "step": 2713
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6130322102924844e-05,
      "loss": 1.4848,
      "step": 2714
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6128841169937062e-05,
      "loss": 1.3728,
      "step": 2715
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.612736023694928e-05,
      "loss": 1.698,
      "step": 2716
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6125879303961497e-05,
      "loss": 1.6128,
      "step": 2717
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6124398370973714e-05,
      "loss": 1.802,
      "step": 2718
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6122917437985932e-05,
      "loss": 1.6366,
      "step": 2719
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.612143650499815e-05,
      "loss": 1.464,
      "step": 2720
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6119955572010367e-05,
      "loss": 1.5413,
      "step": 2721
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6118474639022584e-05,
      "loss": 1.3331,
      "step": 2722
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6116993706034802e-05,
      "loss": 1.343,
      "step": 2723
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6115512773047023e-05,
      "loss": 1.2519,
      "step": 2724
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.611403184005924e-05,
      "loss": 1.3449,
      "step": 2725
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6112550907071454e-05,
      "loss": 1.3093,
      "step": 2726
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6111069974083675e-05,
      "loss": 1.4186,
      "step": 2727
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6109589041095893e-05,
      "loss": 1.5552,
      "step": 2728
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.610810810810811e-05,
      "loss": 1.4467,
      "step": 2729
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6106627175120328e-05,
      "loss": 1.6824,
      "step": 2730
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6105146242132545e-05,
      "loss": 1.4871,
      "step": 2731
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6103665309144763e-05,
      "loss": 1.6335,
      "step": 2732
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.610218437615698e-05,
      "loss": 1.466,
      "step": 2733
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6100703443169198e-05,
      "loss": 1.3072,
      "step": 2734
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6099222510181415e-05,
      "loss": 1.4174,
      "step": 2735
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6097741577193633e-05,
      "loss": 1.423,
      "step": 2736
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.609626064420585e-05,
      "loss": 1.3712,
      "step": 2737
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6094779711218068e-05,
      "loss": 1.4938,
      "step": 2738
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6093298778230286e-05,
      "loss": 1.295,
      "step": 2739
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6091817845242503e-05,
      "loss": 1.4249,
      "step": 2740
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6090336912254724e-05,
      "loss": 1.5072,
      "step": 2741
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6088855979266938e-05,
      "loss": 1.4944,
      "step": 2742
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.6087375046279156e-05,
      "loss": 1.3738,
      "step": 2743
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6085894113291377e-05,
      "loss": 1.4208,
      "step": 2744
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6084413180303594e-05,
      "loss": 1.3657,
      "step": 2745
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.608293224731581e-05,
      "loss": 1.5357,
      "step": 2746
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6081451314328026e-05,
      "loss": 1.3294,
      "step": 2747
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6079970381340247e-05,
      "loss": 1.4007,
      "step": 2748
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6078489448352464e-05,
      "loss": 1.4567,
      "step": 2749
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.607700851536468e-05,
      "loss": 1.5084,
      "step": 2750
    },
    {
      "epoch": 0.61,
      "eval_loss": NaN,
      "eval_runtime": 4.382,
      "eval_samples_per_second": 352.35,
      "eval_steps_per_second": 44.044,
      "step": 2750
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.60755275823769e-05,
      "loss": 1.1903,
      "step": 2751
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6074046649389117e-05,
      "loss": 1.4024,
      "step": 2752
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6072565716401334e-05,
      "loss": 1.2545,
      "step": 2753
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.607108478341355e-05,
      "loss": 1.6098,
      "step": 2754
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.606960385042577e-05,
      "loss": 1.7568,
      "step": 2755
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6068122917437987e-05,
      "loss": 1.4187,
      "step": 2756
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6066641984450204e-05,
      "loss": 1.3957,
      "step": 2757
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6065161051462422e-05,
      "loss": 1.516,
      "step": 2758
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.606368011847464e-05,
      "loss": 1.6029,
      "step": 2759
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6062199185486857e-05,
      "loss": 1.5344,
      "step": 2760
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6060718252499078e-05,
      "loss": 1.4483,
      "step": 2761
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6059237319511295e-05,
      "loss": 1.5119,
      "step": 2762
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.605775638652351e-05,
      "loss": 1.4336,
      "step": 2763
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6056275453535727e-05,
      "loss": 1.4158,
      "step": 2764
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6054794520547948e-05,
      "loss": 1.4525,
      "step": 2765
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6053313587560165e-05,
      "loss": 1.6164,
      "step": 2766
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6051832654572383e-05,
      "loss": 1.4087,
      "step": 2767
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.60503517215846e-05,
      "loss": 1.359,
      "step": 2768
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6048870788596818e-05,
      "loss": 1.3488,
      "step": 2769
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6047389855609035e-05,
      "loss": 1.5831,
      "step": 2770
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6045908922621253e-05,
      "loss": 1.4463,
      "step": 2771
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.604442798963347e-05,
      "loss": 1.5726,
      "step": 2772
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6042947056645688e-05,
      "loss": 1.404,
      "step": 2773
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6041466123657905e-05,
      "loss": 1.3565,
      "step": 2774
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6039985190670123e-05,
      "loss": 1.4588,
      "step": 2775
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.603850425768234e-05,
      "loss": 1.6734,
      "step": 2776
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6037023324694558e-05,
      "loss": 1.5544,
      "step": 2777
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.603554239170678e-05,
      "loss": 1.6568,
      "step": 2778
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6034061458718993e-05,
      "loss": 1.2503,
      "step": 2779
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.603258052573121e-05,
      "loss": 1.3966,
      "step": 2780
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6031099592743428e-05,
      "loss": 1.4366,
      "step": 2781
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.602961865975565e-05,
      "loss": 1.5795,
      "step": 2782
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6028137726767866e-05,
      "loss": 1.3939,
      "step": 2783
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.602665679378008e-05,
      "loss": 1.6702,
      "step": 2784
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.60251758607923e-05,
      "loss": 1.1628,
      "step": 2785
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.602369492780452e-05,
      "loss": 1.2144,
      "step": 2786
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6022213994816737e-05,
      "loss": 1.559,
      "step": 2787
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.6020733061828954e-05,
      "loss": 1.3741,
      "step": 2788
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.601925212884117e-05,
      "loss": 1.508,
      "step": 2789
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.601777119585339e-05,
      "loss": 1.3378,
      "step": 2790
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6016290262865607e-05,
      "loss": 1.588,
      "step": 2791
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6014809329877824e-05,
      "loss": 1.314,
      "step": 2792
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.601332839689004e-05,
      "loss": 1.2908,
      "step": 2793
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.601184746390226e-05,
      "loss": 1.35,
      "step": 2794
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.601036653091448e-05,
      "loss": 1.6264,
      "step": 2795
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6008885597926694e-05,
      "loss": 1.258,
      "step": 2796
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.600740466493891e-05,
      "loss": 1.4444,
      "step": 2797
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.600592373195113e-05,
      "loss": 1.3319,
      "step": 2798
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.600444279896335e-05,
      "loss": 1.3633,
      "step": 2799
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6002961865975564e-05,
      "loss": 1.2471,
      "step": 2800
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6001480932987782e-05,
      "loss": 1.4007,
      "step": 2801
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.3973,
      "step": 2802
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.599851906701222e-05,
      "loss": 1.3356,
      "step": 2803
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5997038134024438e-05,
      "loss": 1.5106,
      "step": 2804
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5995557201036652e-05,
      "loss": 1.4476,
      "step": 2805
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5994076268048873e-05,
      "loss": 1.7186,
      "step": 2806
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.599259533506109e-05,
      "loss": 1.5382,
      "step": 2807
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5991114402073308e-05,
      "loss": 1.452,
      "step": 2808
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5989633469085525e-05,
      "loss": 1.519,
      "step": 2809
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5988152536097743e-05,
      "loss": 1.2921,
      "step": 2810
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.598667160310996e-05,
      "loss": 1.43,
      "step": 2811
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5985190670122178e-05,
      "loss": 1.7562,
      "step": 2812
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5983709737134395e-05,
      "loss": 1.5616,
      "step": 2813
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5982228804146613e-05,
      "loss": 1.3962,
      "step": 2814
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.598074787115883e-05,
      "loss": 1.4186,
      "step": 2815
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.597926693817105e-05,
      "loss": 1.4819,
      "step": 2816
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5977786005183265e-05,
      "loss": 1.4858,
      "step": 2817
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5976305072195483e-05,
      "loss": 1.3873,
      "step": 2818
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5974824139207704e-05,
      "loss": 1.4944,
      "step": 2819
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.597334320621992e-05,
      "loss": 1.3862,
      "step": 2820
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5971862273232135e-05,
      "loss": 1.569,
      "step": 2821
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5970381340244353e-05,
      "loss": 1.3706,
      "step": 2822
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5968900407256574e-05,
      "loss": 1.4413,
      "step": 2823
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.596741947426879e-05,
      "loss": 1.2414,
      "step": 2824
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.596593854128101e-05,
      "loss": 1.3985,
      "step": 2825
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5964457608293226e-05,
      "loss": 1.6316,
      "step": 2826
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5962976675305444e-05,
      "loss": 1.3331,
      "step": 2827
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.596149574231766e-05,
      "loss": 1.5322,
      "step": 2828
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.596001480932988e-05,
      "loss": 1.3598,
      "step": 2829
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5958533876342097e-05,
      "loss": 1.4611,
      "step": 2830
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5957052943354314e-05,
      "loss": 1.1567,
      "step": 2831
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.595557201036653e-05,
      "loss": 1.6987,
      "step": 2832
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.595409107737875e-05,
      "loss": 1.5092,
      "step": 2833
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.5952610144390967e-05,
      "loss": 1.3449,
      "step": 2834
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5951129211403184e-05,
      "loss": 1.5006,
      "step": 2835
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5949648278415405e-05,
      "loss": 1.4957,
      "step": 2836
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5948167345427623e-05,
      "loss": 1.2156,
      "step": 2837
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5946686412439837e-05,
      "loss": 1.4404,
      "step": 2838
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5945205479452054e-05,
      "loss": 1.4121,
      "step": 2839
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5943724546464275e-05,
      "loss": 1.5306,
      "step": 2840
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5942243613476493e-05,
      "loss": 1.3884,
      "step": 2841
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5940762680488707e-05,
      "loss": 1.3927,
      "step": 2842
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5939281747500928e-05,
      "loss": 1.407,
      "step": 2843
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5937800814513145e-05,
      "loss": 1.2578,
      "step": 2844
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5936319881525363e-05,
      "loss": 1.4633,
      "step": 2845
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.593483894853758e-05,
      "loss": 1.442,
      "step": 2846
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5933358015549798e-05,
      "loss": 1.3273,
      "step": 2847
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5931877082562015e-05,
      "loss": 1.4287,
      "step": 2848
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5930396149574233e-05,
      "loss": 1.3519,
      "step": 2849
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.592891521658645e-05,
      "loss": 1.4889,
      "step": 2850
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5927434283598668e-05,
      "loss": 1.2255,
      "step": 2851
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5925953350610885e-05,
      "loss": 1.4957,
      "step": 2852
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5924472417623106e-05,
      "loss": 1.4262,
      "step": 2853
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.592299148463532e-05,
      "loss": 1.4363,
      "step": 2854
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5921510551647538e-05,
      "loss": 1.4801,
      "step": 2855
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.592002961865976e-05,
      "loss": 1.1027,
      "step": 2856
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5918548685671976e-05,
      "loss": 1.3553,
      "step": 2857
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5917067752684194e-05,
      "loss": 1.3111,
      "step": 2858
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5915586819696408e-05,
      "loss": 1.4368,
      "step": 2859
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.591410588670863e-05,
      "loss": 1.5448,
      "step": 2860
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5912624953720846e-05,
      "loss": 1.3935,
      "step": 2861
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5911144020733064e-05,
      "loss": 1.2912,
      "step": 2862
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.590966308774528e-05,
      "loss": 1.3139,
      "step": 2863
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.59081821547575e-05,
      "loss": 1.313,
      "step": 2864
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5906701221769716e-05,
      "loss": 1.476,
      "step": 2865
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5905220288781934e-05,
      "loss": 1.2873,
      "step": 2866
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.590373935579415e-05,
      "loss": 1.302,
      "step": 2867
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.590225842280637e-05,
      "loss": 1.6625,
      "step": 2868
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5900777489818586e-05,
      "loss": 1.4981,
      "step": 2869
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5899296556830804e-05,
      "loss": 1.4868,
      "step": 2870
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.589781562384302e-05,
      "loss": 1.3032,
      "step": 2871
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.589633469085524e-05,
      "loss": 1.378,
      "step": 2872
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.589485375786746e-05,
      "loss": 1.3282,
      "step": 2873
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5893372824879677e-05,
      "loss": 1.5129,
      "step": 2874
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.589189189189189e-05,
      "loss": 1.3602,
      "step": 2875
    },
    {
      "epoch": 0.63,
      "eval_loss": NaN,
      "eval_runtime": 4.3728,
      "eval_samples_per_second": 353.093,
      "eval_steps_per_second": 44.137,
      "step": 2875
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.589041095890411e-05,
      "loss": 1.5322,
      "step": 2876
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.588893002591633e-05,
      "loss": 1.464,
      "step": 2877
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5887449092928548e-05,
      "loss": 1.5342,
      "step": 2878
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.5885968159940765e-05,
      "loss": 1.4919,
      "step": 2879
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5884487226952983e-05,
      "loss": 1.2483,
      "step": 2880
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.58830062939652e-05,
      "loss": 1.4625,
      "step": 2881
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5881525360977418e-05,
      "loss": 1.6148,
      "step": 2882
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5880044427989635e-05,
      "loss": 1.6348,
      "step": 2883
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5878563495001853e-05,
      "loss": 1.5387,
      "step": 2884
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.587708256201407e-05,
      "loss": 1.2107,
      "step": 2885
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5875601629026288e-05,
      "loss": 1.4289,
      "step": 2886
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5874120696038505e-05,
      "loss": 1.3644,
      "step": 2887
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5872639763050723e-05,
      "loss": 1.5779,
      "step": 2888
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.587115883006294e-05,
      "loss": 1.497,
      "step": 2889
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.586967789707516e-05,
      "loss": 1.4573,
      "step": 2890
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5868196964087375e-05,
      "loss": 1.4721,
      "step": 2891
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5866716031099593e-05,
      "loss": 1.6064,
      "step": 2892
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.586523509811181e-05,
      "loss": 1.5576,
      "step": 2893
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.586375416512403e-05,
      "loss": 1.3925,
      "step": 2894
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.586227323213625e-05,
      "loss": 1.4552,
      "step": 2895
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5860792299148463e-05,
      "loss": 1.3683,
      "step": 2896
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5859311366160684e-05,
      "loss": 1.32,
      "step": 2897
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.58578304331729e-05,
      "loss": 1.6672,
      "step": 2898
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.585634950018512e-05,
      "loss": 1.3765,
      "step": 2899
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5854868567197336e-05,
      "loss": 1.6126,
      "step": 2900
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5853387634209554e-05,
      "loss": 1.4724,
      "step": 2901
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.585190670122177e-05,
      "loss": 1.5078,
      "step": 2902
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.585042576823399e-05,
      "loss": 1.5675,
      "step": 2903
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5848944835246206e-05,
      "loss": 1.4426,
      "step": 2904
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5847463902258424e-05,
      "loss": 1.4425,
      "step": 2905
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.584598296927064e-05,
      "loss": 1.5987,
      "step": 2906
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.584450203628286e-05,
      "loss": 1.4602,
      "step": 2907
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5843021103295076e-05,
      "loss": 1.408,
      "step": 2908
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5841540170307294e-05,
      "loss": 1.3518,
      "step": 2909
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.584005923731951e-05,
      "loss": 1.5049,
      "step": 2910
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5838578304331732e-05,
      "loss": 1.4931,
      "step": 2911
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5837097371343946e-05,
      "loss": 1.6987,
      "step": 2912
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5835616438356164e-05,
      "loss": 1.4152,
      "step": 2913
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5834135505368385e-05,
      "loss": 1.5094,
      "step": 2914
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5832654572380602e-05,
      "loss": 1.3964,
      "step": 2915
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.583117363939282e-05,
      "loss": 1.5856,
      "step": 2916
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5829692706405034e-05,
      "loss": 1.5676,
      "step": 2917
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5828211773417255e-05,
      "loss": 1.3915,
      "step": 2918
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5826730840429472e-05,
      "loss": 1.6075,
      "step": 2919
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.582524990744169e-05,
      "loss": 1.346,
      "step": 2920
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5823768974453908e-05,
      "loss": 1.5366,
      "step": 2921
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5822288041466125e-05,
      "loss": 1.3665,
      "step": 2922
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5820807108478343e-05,
      "loss": 1.2881,
      "step": 2923
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.581932617549056e-05,
      "loss": 1.611,
      "step": 2924
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.5817845242502778e-05,
      "loss": 1.4117,
      "step": 2925
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5816364309514995e-05,
      "loss": 1.6745,
      "step": 2926
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5814883376527213e-05,
      "loss": 1.2673,
      "step": 2927
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.581340244353943e-05,
      "loss": 1.3569,
      "step": 2928
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5811921510551648e-05,
      "loss": 1.3454,
      "step": 2929
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5810440577563865e-05,
      "loss": 1.3699,
      "step": 2930
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5808959644576086e-05,
      "loss": 1.4712,
      "step": 2931
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5807478711588304e-05,
      "loss": 1.3647,
      "step": 2932
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5805997778600518e-05,
      "loss": 1.4473,
      "step": 2933
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5804516845612735e-05,
      "loss": 1.6117,
      "step": 2934
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5803035912624956e-05,
      "loss": 1.3876,
      "step": 2935
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5801554979637174e-05,
      "loss": 1.5102,
      "step": 2936
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.580007404664939e-05,
      "loss": 1.451,
      "step": 2937
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.579859311366161e-05,
      "loss": 1.4778,
      "step": 2938
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5797112180673826e-05,
      "loss": 1.2354,
      "step": 2939
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5795631247686044e-05,
      "loss": 1.6172,
      "step": 2940
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.579415031469826e-05,
      "loss": 1.2553,
      "step": 2941
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.579266938171048e-05,
      "loss": 1.6799,
      "step": 2942
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5791188448722696e-05,
      "loss": 1.4501,
      "step": 2943
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5789707515734914e-05,
      "loss": 1.2945,
      "step": 2944
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.578822658274713e-05,
      "loss": 1.4344,
      "step": 2945
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.578674564975935e-05,
      "loss": 1.4025,
      "step": 2946
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5785264716771566e-05,
      "loss": 1.2523,
      "step": 2947
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5783783783783787e-05,
      "loss": 1.6603,
      "step": 2948
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5782302850796e-05,
      "loss": 1.4222,
      "step": 2949
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.578082191780822e-05,
      "loss": 1.2663,
      "step": 2950
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5779340984820436e-05,
      "loss": 1.5772,
      "step": 2951
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5777860051832657e-05,
      "loss": 1.5044,
      "step": 2952
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5776379118844875e-05,
      "loss": 1.36,
      "step": 2953
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.577489818585709e-05,
      "loss": 1.5502,
      "step": 2954
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.577341725286931e-05,
      "loss": 1.4469,
      "step": 2955
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5771936319881527e-05,
      "loss": 1.2773,
      "step": 2956
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5770455386893745e-05,
      "loss": 1.5862,
      "step": 2957
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5768974453905962e-05,
      "loss": 1.4758,
      "step": 2958
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.576749352091818e-05,
      "loss": 1.4324,
      "step": 2959
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5766012587930397e-05,
      "loss": 1.2915,
      "step": 2960
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5764531654942615e-05,
      "loss": 1.314,
      "step": 2961
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5763050721954832e-05,
      "loss": 1.3883,
      "step": 2962
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.576156978896705e-05,
      "loss": 1.4193,
      "step": 2963
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5760088855979268e-05,
      "loss": 1.4779,
      "step": 2964
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.575860792299149e-05,
      "loss": 1.368,
      "step": 2965
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5757126990003703e-05,
      "loss": 1.32,
      "step": 2966
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.575564605701592e-05,
      "loss": 1.463,
      "step": 2967
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5754165124028138e-05,
      "loss": 1.1924,
      "step": 2968
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.575268419104036e-05,
      "loss": 1.4591,
      "step": 2969
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.5751203258052573e-05,
      "loss": 1.5412,
      "step": 2970
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.574972232506479e-05,
      "loss": 1.4454,
      "step": 2971
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.574824139207701e-05,
      "loss": 1.2971,
      "step": 2972
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.574676045908923e-05,
      "loss": 1.5192,
      "step": 2973
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5745279526101446e-05,
      "loss": 1.4202,
      "step": 2974
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5743798593113664e-05,
      "loss": 1.4538,
      "step": 2975
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.574231766012588e-05,
      "loss": 1.5363,
      "step": 2976
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.57408367271381e-05,
      "loss": 1.4392,
      "step": 2977
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5739355794150316e-05,
      "loss": 1.5414,
      "step": 2978
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5737874861162534e-05,
      "loss": 1.4372,
      "step": 2979
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.573639392817475e-05,
      "loss": 1.2557,
      "step": 2980
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.573491299518697e-05,
      "loss": 1.579,
      "step": 2981
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5733432062199186e-05,
      "loss": 1.645,
      "step": 2982
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5731951129211404e-05,
      "loss": 1.3192,
      "step": 2983
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.573047019622362e-05,
      "loss": 1.304,
      "step": 2984
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5728989263235842e-05,
      "loss": 1.4234,
      "step": 2985
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.572750833024806e-05,
      "loss": 1.2194,
      "step": 2986
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5726027397260274e-05,
      "loss": 1.3371,
      "step": 2987
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.572454646427249e-05,
      "loss": 1.515,
      "step": 2988
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5723065531284712e-05,
      "loss": 1.5387,
      "step": 2989
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.572158459829693e-05,
      "loss": 1.5108,
      "step": 2990
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5720103665309144e-05,
      "loss": 1.3021,
      "step": 2991
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5718622732321365e-05,
      "loss": 1.2322,
      "step": 2992
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5717141799333582e-05,
      "loss": 1.81,
      "step": 2993
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.57156608663458e-05,
      "loss": 1.314,
      "step": 2994
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5714179933358017e-05,
      "loss": 1.4413,
      "step": 2995
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5712699000370235e-05,
      "loss": 1.4688,
      "step": 2996
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5711218067382452e-05,
      "loss": 1.5502,
      "step": 2997
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.570973713439467e-05,
      "loss": 1.5199,
      "step": 2998
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5708256201406887e-05,
      "loss": 1.442,
      "step": 2999
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5706775268419105e-05,
      "loss": 1.232,
      "step": 3000
    },
    {
      "epoch": 0.66,
      "eval_loss": NaN,
      "eval_runtime": 4.3735,
      "eval_samples_per_second": 353.032,
      "eval_steps_per_second": 44.129,
      "step": 3000
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5705294335431322e-05,
      "loss": 1.3623,
      "step": 3001
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5703813402443543e-05,
      "loss": 1.3515,
      "step": 3002
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5702332469455757e-05,
      "loss": 1.5328,
      "step": 3003
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5700851536467975e-05,
      "loss": 1.3158,
      "step": 3004
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5699370603480192e-05,
      "loss": 1.5296,
      "step": 3005
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5697889670492413e-05,
      "loss": 1.5796,
      "step": 3006
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.569640873750463e-05,
      "loss": 1.3466,
      "step": 3007
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5694927804516845e-05,
      "loss": 1.3581,
      "step": 3008
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5693446871529066e-05,
      "loss": 1.4989,
      "step": 3009
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5691965938541283e-05,
      "loss": 1.116,
      "step": 3010
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.56904850055535e-05,
      "loss": 1.4405,
      "step": 3011
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5689004072565715e-05,
      "loss": 1.6113,
      "step": 3012
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5687523139577936e-05,
      "loss": 1.144,
      "step": 3013
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5686042206590154e-05,
      "loss": 1.3208,
      "step": 3014
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.568456127360237e-05,
      "loss": 1.3517,
      "step": 3015
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.568308034061459e-05,
      "loss": 1.6946,
      "step": 3016
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5681599407626806e-05,
      "loss": 1.5603,
      "step": 3017
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5680118474639024e-05,
      "loss": 1.3031,
      "step": 3018
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.567863754165124e-05,
      "loss": 1.598,
      "step": 3019
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.567715660866346e-05,
      "loss": 1.3705,
      "step": 3020
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5675675675675676e-05,
      "loss": 1.1519,
      "step": 3021
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5674194742687894e-05,
      "loss": 1.3408,
      "step": 3022
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5672713809700115e-05,
      "loss": 1.4055,
      "step": 3023
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.567123287671233e-05,
      "loss": 1.5093,
      "step": 3024
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5669751943724546e-05,
      "loss": 1.4778,
      "step": 3025
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5668271010736767e-05,
      "loss": 1.3988,
      "step": 3026
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5666790077748985e-05,
      "loss": 1.6816,
      "step": 3027
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5665309144761202e-05,
      "loss": 1.3332,
      "step": 3028
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5663828211773416e-05,
      "loss": 1.3894,
      "step": 3029
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5662347278785637e-05,
      "loss": 1.4078,
      "step": 3030
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5660866345797855e-05,
      "loss": 1.3698,
      "step": 3031
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5659385412810072e-05,
      "loss": 1.6412,
      "step": 3032
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.565790447982229e-05,
      "loss": 1.2349,
      "step": 3033
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5656423546834507e-05,
      "loss": 1.7255,
      "step": 3034
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5654942613846725e-05,
      "loss": 1.7015,
      "step": 3035
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5653461680858942e-05,
      "loss": 1.2723,
      "step": 3036
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.565198074787116e-05,
      "loss": 1.3178,
      "step": 3037
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5650499814883377e-05,
      "loss": 1.3603,
      "step": 3038
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5649018881895595e-05,
      "loss": 1.3731,
      "step": 3039
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5647537948907812e-05,
      "loss": 1.5931,
      "step": 3040
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.564605701592003e-05,
      "loss": 1.4848,
      "step": 3041
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5644576082932247e-05,
      "loss": 1.3462,
      "step": 3042
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5643095149944468e-05,
      "loss": 1.4601,
      "step": 3043
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5641614216956686e-05,
      "loss": 1.3617,
      "step": 3044
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.56401332839689e-05,
      "loss": 1.4583,
      "step": 3045
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5638652350981117e-05,
      "loss": 1.4145,
      "step": 3046
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.563717141799334e-05,
      "loss": 1.5364,
      "step": 3047
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5635690485005556e-05,
      "loss": 1.5978,
      "step": 3048
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.563420955201777e-05,
      "loss": 1.4755,
      "step": 3049
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.563272861902999e-05,
      "loss": 1.3124,
      "step": 3050
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.563124768604221e-05,
      "loss": 1.4012,
      "step": 3051
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5629766753054426e-05,
      "loss": 1.4373,
      "step": 3052
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5628285820066643e-05,
      "loss": 1.3538,
      "step": 3053
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.562680488707886e-05,
      "loss": 1.6602,
      "step": 3054
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.562532395409108e-05,
      "loss": 1.405,
      "step": 3055
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5623843021103296e-05,
      "loss": 1.3112,
      "step": 3056
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5622362088115514e-05,
      "loss": 1.3808,
      "step": 3057
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.562088115512773e-05,
      "loss": 1.2824,
      "step": 3058
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.561940022213995e-05,
      "loss": 1.488,
      "step": 3059
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.561791928915217e-05,
      "loss": 1.3665,
      "step": 3060
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.5616438356164384e-05,
      "loss": 1.202,
      "step": 3061
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.56149574231766e-05,
      "loss": 1.2726,
      "step": 3062
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.561347649018882e-05,
      "loss": 1.4568,
      "step": 3063
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.561199555720104e-05,
      "loss": 1.4277,
      "step": 3064
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5610514624213257e-05,
      "loss": 1.4047,
      "step": 3065
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.560903369122547e-05,
      "loss": 1.3627,
      "step": 3066
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5607552758237692e-05,
      "loss": 1.4345,
      "step": 3067
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.560607182524991e-05,
      "loss": 1.6213,
      "step": 3068
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5604590892262127e-05,
      "loss": 1.2931,
      "step": 3069
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5603109959274345e-05,
      "loss": 1.5662,
      "step": 3070
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5601629026286562e-05,
      "loss": 1.4439,
      "step": 3071
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.560014809329878e-05,
      "loss": 1.4272,
      "step": 3072
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5598667160310997e-05,
      "loss": 1.5551,
      "step": 3073
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5597186227323215e-05,
      "loss": 1.4224,
      "step": 3074
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5595705294335432e-05,
      "loss": 1.3466,
      "step": 3075
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.559422436134765e-05,
      "loss": 1.3238,
      "step": 3076
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5592743428359867e-05,
      "loss": 1.5195,
      "step": 3077
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5591262495372085e-05,
      "loss": 1.2081,
      "step": 3078
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5589781562384302e-05,
      "loss": 1.5327,
      "step": 3079
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.558830062939652e-05,
      "loss": 1.2866,
      "step": 3080
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.558681969640874e-05,
      "loss": 1.5713,
      "step": 3081
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5585338763420955e-05,
      "loss": 1.411,
      "step": 3082
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5583857830433172e-05,
      "loss": 1.5673,
      "step": 3083
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5582376897445393e-05,
      "loss": 1.232,
      "step": 3084
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.558089596445761e-05,
      "loss": 1.4987,
      "step": 3085
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5579415031469828e-05,
      "loss": 1.105,
      "step": 3086
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5577934098482046e-05,
      "loss": 1.2874,
      "step": 3087
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5576453165494263e-05,
      "loss": 1.4742,
      "step": 3088
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.557497223250648e-05,
      "loss": 1.3526,
      "step": 3089
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.55734912995187e-05,
      "loss": 1.4552,
      "step": 3090
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5572010366530916e-05,
      "loss": 1.4045,
      "step": 3091
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5570529433543133e-05,
      "loss": 1.4478,
      "step": 3092
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.556904850055535e-05,
      "loss": 1.2798,
      "step": 3093
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.556756756756757e-05,
      "loss": 1.4207,
      "step": 3094
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5566086634579786e-05,
      "loss": 1.3599,
      "step": 3095
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5564605701592003e-05,
      "loss": 1.3955,
      "step": 3096
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.556312476860422e-05,
      "loss": 1.3373,
      "step": 3097
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.556164383561644e-05,
      "loss": 1.6275,
      "step": 3098
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5560162902628656e-05,
      "loss": 1.6731,
      "step": 3099
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5558681969640874e-05,
      "loss": 1.5598,
      "step": 3100
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5557201036653094e-05,
      "loss": 1.6193,
      "step": 3101
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5555720103665312e-05,
      "loss": 1.3779,
      "step": 3102
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5554239170677526e-05,
      "loss": 1.2867,
      "step": 3103
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5552758237689747e-05,
      "loss": 1.3935,
      "step": 3104
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5551277304701964e-05,
      "loss": 1.4694,
      "step": 3105
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.5549796371714182e-05,
      "loss": 1.3831,
      "step": 3106
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.55483154387264e-05,
      "loss": 1.2417,
      "step": 3107
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5546834505738617e-05,
      "loss": 1.3068,
      "step": 3108
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5545353572750835e-05,
      "loss": 1.4824,
      "step": 3109
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5543872639763052e-05,
      "loss": 1.5258,
      "step": 3110
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.554239170677527e-05,
      "loss": 1.3837,
      "step": 3111
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5540910773787487e-05,
      "loss": 1.3242,
      "step": 3112
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5539429840799705e-05,
      "loss": 1.5465,
      "step": 3113
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5537948907811926e-05,
      "loss": 1.5254,
      "step": 3114
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.553646797482414e-05,
      "loss": 1.2596,
      "step": 3115
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5534987041836357e-05,
      "loss": 1.4369,
      "step": 3116
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5533506108848575e-05,
      "loss": 1.4351,
      "step": 3117
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5532025175860796e-05,
      "loss": 1.3188,
      "step": 3118
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.553054424287301e-05,
      "loss": 1.5266,
      "step": 3119
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5529063309885227e-05,
      "loss": 1.5493,
      "step": 3120
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5527582376897448e-05,
      "loss": 1.3404,
      "step": 3121
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5526101443909666e-05,
      "loss": 1.4639,
      "step": 3122
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5524620510921883e-05,
      "loss": 1.4048,
      "step": 3123
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5523139577934097e-05,
      "loss": 1.6577,
      "step": 3124
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5521658644946318e-05,
      "loss": 1.2863,
      "step": 3125
    },
    {
      "epoch": 0.69,
      "eval_loss": NaN,
      "eval_runtime": 4.3772,
      "eval_samples_per_second": 352.736,
      "eval_steps_per_second": 44.092,
      "step": 3125
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5520177711958536e-05,
      "loss": 1.4853,
      "step": 3126
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5518696778970753e-05,
      "loss": 1.5253,
      "step": 3127
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.551721584598297e-05,
      "loss": 1.3919,
      "step": 3128
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5515734912995188e-05,
      "loss": 1.3313,
      "step": 3129
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5514253980007406e-05,
      "loss": 1.4774,
      "step": 3130
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5512773047019623e-05,
      "loss": 1.8024,
      "step": 3131
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.551129211403184e-05,
      "loss": 1.4252,
      "step": 3132
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.550981118104406e-05,
      "loss": 1.4249,
      "step": 3133
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5508330248056276e-05,
      "loss": 1.2197,
      "step": 3134
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5506849315068497e-05,
      "loss": 1.4765,
      "step": 3135
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.550536838208071e-05,
      "loss": 1.56,
      "step": 3136
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.550388744909293e-05,
      "loss": 1.5369,
      "step": 3137
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.550240651610515e-05,
      "loss": 1.3864,
      "step": 3138
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5500925583117367e-05,
      "loss": 1.2392,
      "step": 3139
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.549944465012958e-05,
      "loss": 1.7091,
      "step": 3140
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.54979637171418e-05,
      "loss": 1.4008,
      "step": 3141
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.549648278415402e-05,
      "loss": 1.4145,
      "step": 3142
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5495001851166237e-05,
      "loss": 1.304,
      "step": 3143
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5493520918178454e-05,
      "loss": 1.4494,
      "step": 3144
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5492039985190672e-05,
      "loss": 1.483,
      "step": 3145
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.549055905220289e-05,
      "loss": 1.3276,
      "step": 3146
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5489078119215107e-05,
      "loss": 1.4321,
      "step": 3147
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5487597186227324e-05,
      "loss": 1.4919,
      "step": 3148
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5486116253239542e-05,
      "loss": 1.2337,
      "step": 3149
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.548463532025176e-05,
      "loss": 1.6617,
      "step": 3150
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5483154387263977e-05,
      "loss": 1.7034,
      "step": 3151
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5481673454276195e-05,
      "loss": 1.391,
      "step": 3152
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5480192521288412e-05,
      "loss": 1.2131,
      "step": 3153
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.547871158830063e-05,
      "loss": 1.3938,
      "step": 3154
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.547723065531285e-05,
      "loss": 1.4772,
      "step": 3155
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5475749722325068e-05,
      "loss": 1.442,
      "step": 3156
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5474268789337282e-05,
      "loss": 1.3036,
      "step": 3157
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.54727878563495e-05,
      "loss": 1.3999,
      "step": 3158
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.547130692336172e-05,
      "loss": 1.5761,
      "step": 3159
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5469825990373938e-05,
      "loss": 1.4702,
      "step": 3160
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5468345057386152e-05,
      "loss": 1.2813,
      "step": 3161
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5466864124398373e-05,
      "loss": 1.4193,
      "step": 3162
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.546538319141059e-05,
      "loss": 1.3869,
      "step": 3163
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5463902258422808e-05,
      "loss": 1.2419,
      "step": 3164
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5462421325435026e-05,
      "loss": 1.5628,
      "step": 3165
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5460940392447243e-05,
      "loss": 1.8581,
      "step": 3166
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.545945945945946e-05,
      "loss": 1.4734,
      "step": 3167
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5457978526471678e-05,
      "loss": 1.7197,
      "step": 3168
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5456497593483896e-05,
      "loss": 1.2154,
      "step": 3169
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5455016660496113e-05,
      "loss": 1.4079,
      "step": 3170
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.545353572750833e-05,
      "loss": 1.3818,
      "step": 3171
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.545205479452055e-05,
      "loss": 1.2817,
      "step": 3172
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5450573861532766e-05,
      "loss": 1.1248,
      "step": 3173
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5449092928544983e-05,
      "loss": 1.3438,
      "step": 3174
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.54476119955572e-05,
      "loss": 1.4905,
      "step": 3175
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5446131062569422e-05,
      "loss": 1.3308,
      "step": 3176
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5444650129581636e-05,
      "loss": 1.4029,
      "step": 3177
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5443169196593853e-05,
      "loss": 1.5476,
      "step": 3178
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5441688263606074e-05,
      "loss": 1.5056,
      "step": 3179
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5440207330618292e-05,
      "loss": 1.6026,
      "step": 3180
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.543872639763051e-05,
      "loss": 1.3318,
      "step": 3181
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5437245464642727e-05,
      "loss": 1.3622,
      "step": 3182
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5435764531654944e-05,
      "loss": 1.354,
      "step": 3183
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5434283598667162e-05,
      "loss": 1.573,
      "step": 3184
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.543280266567938e-05,
      "loss": 1.3662,
      "step": 3185
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5431321732691597e-05,
      "loss": 1.3012,
      "step": 3186
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5429840799703814e-05,
      "loss": 1.4192,
      "step": 3187
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5428359866716032e-05,
      "loss": 1.5062,
      "step": 3188
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.542687893372825e-05,
      "loss": 1.4063,
      "step": 3189
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5425398000740467e-05,
      "loss": 1.2175,
      "step": 3190
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5423917067752684e-05,
      "loss": 1.5628,
      "step": 3191
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5422436134764902e-05,
      "loss": 1.4596,
      "step": 3192
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5420955201777123e-05,
      "loss": 1.7319,
      "step": 3193
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5419474268789337e-05,
      "loss": 1.6056,
      "step": 3194
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5417993335801555e-05,
      "loss": 1.3366,
      "step": 3195
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5416512402813775e-05,
      "loss": 1.3836,
      "step": 3196
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.5415031469825993e-05,
      "loss": 1.5453,
      "step": 3197
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5413550536838207e-05,
      "loss": 1.6435,
      "step": 3198
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5412069603850428e-05,
      "loss": 1.6413,
      "step": 3199
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5410588670862646e-05,
      "loss": 1.453,
      "step": 3200
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5409107737874863e-05,
      "loss": 1.4202,
      "step": 3201
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.540762680488708e-05,
      "loss": 1.2863,
      "step": 3202
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5406145871899298e-05,
      "loss": 1.29,
      "step": 3203
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5404664938911516e-05,
      "loss": 1.3573,
      "step": 3204
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5403184005923733e-05,
      "loss": 1.6094,
      "step": 3205
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.540170307293595e-05,
      "loss": 1.2912,
      "step": 3206
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5400222139948168e-05,
      "loss": 1.7137,
      "step": 3207
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5398741206960386e-05,
      "loss": 1.4995,
      "step": 3208
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5397260273972603e-05,
      "loss": 1.485,
      "step": 3209
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.539577934098482e-05,
      "loss": 1.4827,
      "step": 3210
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5394298407997038e-05,
      "loss": 1.5685,
      "step": 3211
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5392817475009256e-05,
      "loss": 1.5158,
      "step": 3212
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5391336542021477e-05,
      "loss": 1.4198,
      "step": 3213
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5389855609033694e-05,
      "loss": 1.3571,
      "step": 3214
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5388374676045908e-05,
      "loss": 1.4157,
      "step": 3215
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.538689374305813e-05,
      "loss": 1.5811,
      "step": 3216
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5385412810070347e-05,
      "loss": 1.3791,
      "step": 3217
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5383931877082564e-05,
      "loss": 1.4353,
      "step": 3218
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.538245094409478e-05,
      "loss": 1.2609,
      "step": 3219
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5380970011107e-05,
      "loss": 1.2595,
      "step": 3220
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5379489078119217e-05,
      "loss": 1.4207,
      "step": 3221
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5378008145131434e-05,
      "loss": 1.1829,
      "step": 3222
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5376527212143652e-05,
      "loss": 1.383,
      "step": 3223
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.537504627915587e-05,
      "loss": 1.4311,
      "step": 3224
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5373565346168087e-05,
      "loss": 1.7399,
      "step": 3225
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5372084413180304e-05,
      "loss": 1.1106,
      "step": 3226
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5370603480192522e-05,
      "loss": 1.4287,
      "step": 3227
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.536912254720474e-05,
      "loss": 1.643,
      "step": 3228
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5367641614216957e-05,
      "loss": 1.3536,
      "step": 3229
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5366160681229178e-05,
      "loss": 1.5224,
      "step": 3230
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5364679748241392e-05,
      "loss": 1.5124,
      "step": 3231
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.536319881525361e-05,
      "loss": 1.4276,
      "step": 3232
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.536171788226583e-05,
      "loss": 1.5117,
      "step": 3233
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5360236949278048e-05,
      "loss": 1.5173,
      "step": 3234
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5358756016290265e-05,
      "loss": 1.1169,
      "step": 3235
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.535727508330248e-05,
      "loss": 1.6189,
      "step": 3236
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.53557941503147e-05,
      "loss": 1.3303,
      "step": 3237
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5354313217326918e-05,
      "loss": 1.4937,
      "step": 3238
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5352832284339135e-05,
      "loss": 1.5476,
      "step": 3239
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5351351351351353e-05,
      "loss": 1.5237,
      "step": 3240
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.534987041836357e-05,
      "loss": 1.4882,
      "step": 3241
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.5348389485375788e-05,
      "loss": 1.6622,
      "step": 3242
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5346908552388006e-05,
      "loss": 1.4869,
      "step": 3243
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5345427619400223e-05,
      "loss": 1.4599,
      "step": 3244
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.534394668641244e-05,
      "loss": 1.4405,
      "step": 3245
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5342465753424658e-05,
      "loss": 1.4023,
      "step": 3246
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5340984820436876e-05,
      "loss": 1.3943,
      "step": 3247
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5339503887449093e-05,
      "loss": 1.4438,
      "step": 3248
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.533802295446131e-05,
      "loss": 1.1782,
      "step": 3249
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.533654202147353e-05,
      "loss": 1.6271,
      "step": 3250
    },
    {
      "epoch": 0.72,
      "eval_loss": NaN,
      "eval_runtime": 4.3804,
      "eval_samples_per_second": 352.482,
      "eval_steps_per_second": 44.06,
      "step": 3250
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.533506108848575e-05,
      "loss": 1.4043,
      "step": 3251
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5333580155497963e-05,
      "loss": 1.4136,
      "step": 3252
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.533209922251018e-05,
      "loss": 1.3044,
      "step": 3253
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.53306182895224e-05,
      "loss": 1.7007,
      "step": 3254
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.532913735653462e-05,
      "loss": 1.3267,
      "step": 3255
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5327656423546837e-05,
      "loss": 1.6382,
      "step": 3256
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5326175490559054e-05,
      "loss": 1.4557,
      "step": 3257
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.532469455757127e-05,
      "loss": 1.4347,
      "step": 3258
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.532321362458349e-05,
      "loss": 1.2916,
      "step": 3259
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5321732691595707e-05,
      "loss": 1.2442,
      "step": 3260
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5320251758607924e-05,
      "loss": 1.3875,
      "step": 3261
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5318770825620142e-05,
      "loss": 1.4377,
      "step": 3262
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.531728989263236e-05,
      "loss": 1.441,
      "step": 3263
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5315808959644577e-05,
      "loss": 1.3945,
      "step": 3264
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5314328026656794e-05,
      "loss": 1.6054,
      "step": 3265
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5312847093669012e-05,
      "loss": 1.5036,
      "step": 3266
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5311366160681233e-05,
      "loss": 1.4155,
      "step": 3267
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5309885227693447e-05,
      "loss": 1.5662,
      "step": 3268
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5308404294705664e-05,
      "loss": 1.3063,
      "step": 3269
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5306923361717882e-05,
      "loss": 1.3571,
      "step": 3270
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5305442428730103e-05,
      "loss": 1.329,
      "step": 3271
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.530396149574232e-05,
      "loss": 1.6138,
      "step": 3272
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5302480562754534e-05,
      "loss": 1.3431,
      "step": 3273
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5300999629766755e-05,
      "loss": 1.4794,
      "step": 3274
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5299518696778973e-05,
      "loss": 1.2497,
      "step": 3275
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.529803776379119e-05,
      "loss": 1.5468,
      "step": 3276
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5296556830803408e-05,
      "loss": 1.3984,
      "step": 3277
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5295075897815625e-05,
      "loss": 1.5143,
      "step": 3278
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5293594964827843e-05,
      "loss": 1.4006,
      "step": 3279
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.529211403184006e-05,
      "loss": 1.5746,
      "step": 3280
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5290633098852278e-05,
      "loss": 1.5351,
      "step": 3281
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5289152165864495e-05,
      "loss": 1.4497,
      "step": 3282
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5287671232876713e-05,
      "loss": 1.3924,
      "step": 3283
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5286190299888934e-05,
      "loss": 1.4192,
      "step": 3284
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5284709366901148e-05,
      "loss": 1.4689,
      "step": 3285
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5283228433913366e-05,
      "loss": 1.6013,
      "step": 3286
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.5281747500925583e-05,
      "loss": 1.394,
      "step": 3287
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5280266567937804e-05,
      "loss": 1.42,
      "step": 3288
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5278785634950018e-05,
      "loss": 1.5139,
      "step": 3289
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5277304701962236e-05,
      "loss": 1.3782,
      "step": 3290
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5275823768974457e-05,
      "loss": 1.5802,
      "step": 3291
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5274342835986674e-05,
      "loss": 1.5114,
      "step": 3292
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.527286190299889e-05,
      "loss": 1.4181,
      "step": 3293
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.527138097001111e-05,
      "loss": 1.6874,
      "step": 3294
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5269900037023327e-05,
      "loss": 1.2951,
      "step": 3295
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5268419104035544e-05,
      "loss": 1.5981,
      "step": 3296
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.526693817104776e-05,
      "loss": 1.4258,
      "step": 3297
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.526545723805998e-05,
      "loss": 1.6429,
      "step": 3298
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5263976305072197e-05,
      "loss": 1.4858,
      "step": 3299
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5262495372084414e-05,
      "loss": 1.3124,
      "step": 3300
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.526101443909663e-05,
      "loss": 1.4725,
      "step": 3301
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.525953350610885e-05,
      "loss": 1.3436,
      "step": 3302
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5258052573121067e-05,
      "loss": 1.4415,
      "step": 3303
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5256571640133284e-05,
      "loss": 1.3399,
      "step": 3304
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5255090707145503e-05,
      "loss": 1.2914,
      "step": 3305
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5253609774157721e-05,
      "loss": 1.3511,
      "step": 3306
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5252128841169937e-05,
      "loss": 1.563,
      "step": 3307
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5250647908182156e-05,
      "loss": 1.4191,
      "step": 3308
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5249166975194374e-05,
      "loss": 1.4166,
      "step": 3309
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5247686042206591e-05,
      "loss": 1.3221,
      "step": 3310
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.524620510921881e-05,
      "loss": 1.5937,
      "step": 3311
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5244724176231028e-05,
      "loss": 1.4821,
      "step": 3312
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5243243243243244e-05,
      "loss": 1.667,
      "step": 3313
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5241762310255461e-05,
      "loss": 1.2662,
      "step": 3314
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.524028137726768e-05,
      "loss": 1.4121,
      "step": 3315
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5238800444279898e-05,
      "loss": 1.5215,
      "step": 3316
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5237319511292115e-05,
      "loss": 1.5042,
      "step": 3317
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5235838578304335e-05,
      "loss": 1.6303,
      "step": 3318
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.523435764531655e-05,
      "loss": 1.4368,
      "step": 3319
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5232876712328768e-05,
      "loss": 1.7726,
      "step": 3320
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5231395779340985e-05,
      "loss": 1.797,
      "step": 3321
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5229914846353205e-05,
      "loss": 1.4794,
      "step": 3322
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5228433913365422e-05,
      "loss": 1.5222,
      "step": 3323
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5226952980377638e-05,
      "loss": 1.6663,
      "step": 3324
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5225472047389857e-05,
      "loss": 1.4473,
      "step": 3325
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5223991114402075e-05,
      "loss": 1.4173,
      "step": 3326
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5222510181414292e-05,
      "loss": 1.365,
      "step": 3327
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5221029248426511e-05,
      "loss": 1.337,
      "step": 3328
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5219548315438727e-05,
      "loss": 1.468,
      "step": 3329
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5218067382450945e-05,
      "loss": 1.1823,
      "step": 3330
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5216586449463162e-05,
      "loss": 1.6371,
      "step": 3331
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5215105516475381e-05,
      "loss": 1.3822,
      "step": 3332
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.5213624583487599e-05,
      "loss": 1.5957,
      "step": 3333
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5212143650499815e-05,
      "loss": 1.3958,
      "step": 3334
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5210662717512034e-05,
      "loss": 1.4225,
      "step": 3335
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5209181784524252e-05,
      "loss": 1.6104,
      "step": 3336
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5207700851536469e-05,
      "loss": 1.7112,
      "step": 3337
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5206219918548687e-05,
      "loss": 1.6103,
      "step": 3338
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5204738985560906e-05,
      "loss": 1.2089,
      "step": 3339
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5203258052573122e-05,
      "loss": 1.3627,
      "step": 3340
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5201777119585339e-05,
      "loss": 1.3917,
      "step": 3341
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5200296186597558e-05,
      "loss": 1.6251,
      "step": 3342
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5198815253609776e-05,
      "loss": 1.2593,
      "step": 3343
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5197334320621993e-05,
      "loss": 1.4213,
      "step": 3344
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5195853387634213e-05,
      "loss": 1.4024,
      "step": 3345
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5194372454646428e-05,
      "loss": 1.3659,
      "step": 3346
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5192891521658646e-05,
      "loss": 1.4347,
      "step": 3347
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5191410588670863e-05,
      "loss": 1.5984,
      "step": 3348
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5189929655683083e-05,
      "loss": 1.5386,
      "step": 3349
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5188448722695298e-05,
      "loss": 1.5769,
      "step": 3350
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5186967789707516e-05,
      "loss": 1.5035,
      "step": 3351
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5185486856719735e-05,
      "loss": 1.5099,
      "step": 3352
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5184005923731953e-05,
      "loss": 1.2809,
      "step": 3353
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.518252499074417e-05,
      "loss": 1.3869,
      "step": 3354
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5181044057756386e-05,
      "loss": 1.5168,
      "step": 3355
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5179563124768605e-05,
      "loss": 1.4112,
      "step": 3356
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5178082191780823e-05,
      "loss": 1.5697,
      "step": 3357
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.517660125879304e-05,
      "loss": 1.1693,
      "step": 3358
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.517512032580526e-05,
      "loss": 1.5287,
      "step": 3359
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5173639392817477e-05,
      "loss": 1.3072,
      "step": 3360
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5172158459829693e-05,
      "loss": 1.45,
      "step": 3361
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5170677526841912e-05,
      "loss": 1.2499,
      "step": 3362
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.516919659385413e-05,
      "loss": 1.1424,
      "step": 3363
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5167715660866347e-05,
      "loss": 1.2818,
      "step": 3364
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5166234727878563e-05,
      "loss": 1.4727,
      "step": 3365
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5164753794890784e-05,
      "loss": 1.3475,
      "step": 3366
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5163272861903e-05,
      "loss": 1.3623,
      "step": 3367
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5161791928915217e-05,
      "loss": 1.2522,
      "step": 3368
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5160310995927436e-05,
      "loss": 1.4287,
      "step": 3369
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5158830062939654e-05,
      "loss": 1.4068,
      "step": 3370
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.515734912995187e-05,
      "loss": 1.4712,
      "step": 3371
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5155868196964087e-05,
      "loss": 1.2491,
      "step": 3372
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5154387263976306e-05,
      "loss": 1.4363,
      "step": 3373
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5152906330988524e-05,
      "loss": 1.3297,
      "step": 3374
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5151425398000741e-05,
      "loss": 1.4499,
      "step": 3375
    },
    {
      "epoch": 0.74,
      "eval_loss": NaN,
      "eval_runtime": 4.3732,
      "eval_samples_per_second": 353.062,
      "eval_steps_per_second": 44.133,
      "step": 3375
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.514994446501296e-05,
      "loss": 1.5515,
      "step": 3376
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5148463532025177e-05,
      "loss": 1.4755,
      "step": 3377
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5146982599037394e-05,
      "loss": 1.5339,
      "step": 3378
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5145501666049613e-05,
      "loss": 1.429,
      "step": 3379
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.514402073306183e-05,
      "loss": 1.5721,
      "step": 3380
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5142539800074048e-05,
      "loss": 1.5312,
      "step": 3381
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5141058867086264e-05,
      "loss": 1.526,
      "step": 3382
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5139577934098483e-05,
      "loss": 1.2985,
      "step": 3383
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.51380970011107e-05,
      "loss": 1.2798,
      "step": 3384
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5136616068122918e-05,
      "loss": 1.4619,
      "step": 3385
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5135135135135138e-05,
      "loss": 1.5268,
      "step": 3386
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5133654202147355e-05,
      "loss": 1.4994,
      "step": 3387
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5132173269159571e-05,
      "loss": 1.2288,
      "step": 3388
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.513069233617179e-05,
      "loss": 1.5566,
      "step": 3389
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5129211403184008e-05,
      "loss": 1.6129,
      "step": 3390
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5127730470196225e-05,
      "loss": 1.5213,
      "step": 3391
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5126249537208441e-05,
      "loss": 1.221,
      "step": 3392
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.512476860422066e-05,
      "loss": 1.3882,
      "step": 3393
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5123287671232878e-05,
      "loss": 1.4765,
      "step": 3394
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5121806738245095e-05,
      "loss": 1.4695,
      "step": 3395
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5120325805257314e-05,
      "loss": 1.4245,
      "step": 3396
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5118844872269532e-05,
      "loss": 1.5168,
      "step": 3397
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5117363939281748e-05,
      "loss": 1.8057,
      "step": 3398
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5115883006293965e-05,
      "loss": 1.3212,
      "step": 3399
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5114402073306184e-05,
      "loss": 1.3943,
      "step": 3400
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5112921140318402e-05,
      "loss": 1.3079,
      "step": 3401
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.511144020733062e-05,
      "loss": 1.4016,
      "step": 3402
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5109959274342839e-05,
      "loss": 1.4292,
      "step": 3403
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5108478341355055e-05,
      "loss": 1.6617,
      "step": 3404
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5106997408367272e-05,
      "loss": 1.3761,
      "step": 3405
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5105516475379491e-05,
      "loss": 1.4171,
      "step": 3406
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5104035542391709e-05,
      "loss": 1.655,
      "step": 3407
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5102554609403926e-05,
      "loss": 1.2173,
      "step": 3408
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5101073676416142e-05,
      "loss": 1.2869,
      "step": 3409
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5099592743428361e-05,
      "loss": 1.2653,
      "step": 3410
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5098111810440579e-05,
      "loss": 1.5756,
      "step": 3411
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5096630877452796e-05,
      "loss": 1.3342,
      "step": 3412
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5095149944465016e-05,
      "loss": 1.6072,
      "step": 3413
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5093669011477231e-05,
      "loss": 1.5159,
      "step": 3414
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5092188078489449e-05,
      "loss": 1.4062,
      "step": 3415
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5090707145501666e-05,
      "loss": 1.6685,
      "step": 3416
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5089226212513886e-05,
      "loss": 1.3024,
      "step": 3417
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5087745279526103e-05,
      "loss": 1.3452,
      "step": 3418
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5086264346538319e-05,
      "loss": 1.3348,
      "step": 3419
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5084783413550538e-05,
      "loss": 1.4776,
      "step": 3420
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5083302480562756e-05,
      "loss": 1.7031,
      "step": 3421
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5081821547574973e-05,
      "loss": 1.4669,
      "step": 3422
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.5080340614587192e-05,
      "loss": 1.5619,
      "step": 3423
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.507885968159941e-05,
      "loss": 1.3592,
      "step": 3424
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5077378748611626e-05,
      "loss": 1.4439,
      "step": 3425
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5075897815623843e-05,
      "loss": 1.5245,
      "step": 3426
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5074416882636063e-05,
      "loss": 1.1976,
      "step": 3427
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.507293594964828e-05,
      "loss": 1.1858,
      "step": 3428
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5071455016660496e-05,
      "loss": 1.3359,
      "step": 3429
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5069974083672717e-05,
      "loss": 1.3968,
      "step": 3430
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5068493150684933e-05,
      "loss": 1.2941,
      "step": 3431
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.506701221769715e-05,
      "loss": 1.5915,
      "step": 3432
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5065531284709368e-05,
      "loss": 1.4662,
      "step": 3433
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5064050351721587e-05,
      "loss": 1.5584,
      "step": 3434
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5062569418733803e-05,
      "loss": 1.4231,
      "step": 3435
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.506108848574602e-05,
      "loss": 1.5749,
      "step": 3436
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.505960755275824e-05,
      "loss": 1.3279,
      "step": 3437
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5058126619770457e-05,
      "loss": 1.3844,
      "step": 3438
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5056645686782674e-05,
      "loss": 1.5588,
      "step": 3439
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5055164753794894e-05,
      "loss": 1.413,
      "step": 3440
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.505368382080711e-05,
      "loss": 1.4033,
      "step": 3441
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5052202887819327e-05,
      "loss": 1.4198,
      "step": 3442
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5050721954831544e-05,
      "loss": 1.4468,
      "step": 3443
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5049241021843764e-05,
      "loss": 1.1877,
      "step": 3444
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5047760088855981e-05,
      "loss": 1.6,
      "step": 3445
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5046279155868197e-05,
      "loss": 1.427,
      "step": 3446
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5044798222880416e-05,
      "loss": 1.4537,
      "step": 3447
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5043317289892634e-05,
      "loss": 1.4873,
      "step": 3448
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5041836356904851e-05,
      "loss": 1.3127,
      "step": 3449
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5040355423917067e-05,
      "loss": 1.329,
      "step": 3450
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5038874490929288e-05,
      "loss": 1.3774,
      "step": 3451
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5037393557941504e-05,
      "loss": 1.4246,
      "step": 3452
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5035912624953721e-05,
      "loss": 1.4035,
      "step": 3453
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.503443169196594e-05,
      "loss": 1.6292,
      "step": 3454
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5032950758978158e-05,
      "loss": 1.4432,
      "step": 3455
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5031469825990374e-05,
      "loss": 1.2682,
      "step": 3456
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5029988893002593e-05,
      "loss": 1.3493,
      "step": 3457
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.502850796001481e-05,
      "loss": 1.2333,
      "step": 3458
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5027027027027028e-05,
      "loss": 1.4538,
      "step": 3459
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5025546094039246e-05,
      "loss": 1.5242,
      "step": 3460
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5024065161051465e-05,
      "loss": 1.4499,
      "step": 3461
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.502258422806368e-05,
      "loss": 1.5458,
      "step": 3462
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5021103295075898e-05,
      "loss": 1.436,
      "step": 3463
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5019622362088117e-05,
      "loss": 1.351,
      "step": 3464
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5018141429100335e-05,
      "loss": 1.3427,
      "step": 3465
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5016660496112552e-05,
      "loss": 1.5205,
      "step": 3466
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5015179563124768e-05,
      "loss": 1.2682,
      "step": 3467
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5013698630136988e-05,
      "loss": 1.6107,
      "step": 3468
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.5012217697149205e-05,
      "loss": 1.3107,
      "step": 3469
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5010736764161423e-05,
      "loss": 1.3307,
      "step": 3470
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5009255831173642e-05,
      "loss": 1.5416,
      "step": 3471
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.500777489818586e-05,
      "loss": 1.3299,
      "step": 3472
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5006293965198075e-05,
      "loss": 1.5601,
      "step": 3473
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5004813032210294e-05,
      "loss": 1.236,
      "step": 3474
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5003332099222512e-05,
      "loss": 1.5663,
      "step": 3475
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.500185116623473e-05,
      "loss": 1.4233,
      "step": 3476
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.5000370233246945e-05,
      "loss": 1.5364,
      "step": 3477
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4998889300259164e-05,
      "loss": 1.5228,
      "step": 3478
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4997408367271382e-05,
      "loss": 1.4386,
      "step": 3479
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.49959274342836e-05,
      "loss": 1.372,
      "step": 3480
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4994446501295819e-05,
      "loss": 1.3415,
      "step": 3481
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4992965568308036e-05,
      "loss": 1.229,
      "step": 3482
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4991484635320252e-05,
      "loss": 1.552,
      "step": 3483
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.499000370233247e-05,
      "loss": 1.4993,
      "step": 3484
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4988522769344689e-05,
      "loss": 1.2131,
      "step": 3485
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4987041836356906e-05,
      "loss": 1.6825,
      "step": 3486
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4985560903369124e-05,
      "loss": 1.2527,
      "step": 3487
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4984079970381343e-05,
      "loss": 1.2864,
      "step": 3488
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4982599037393559e-05,
      "loss": 1.366,
      "step": 3489
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4981118104405776e-05,
      "loss": 1.407,
      "step": 3490
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4979637171417995e-05,
      "loss": 1.5662,
      "step": 3491
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4978156238430213e-05,
      "loss": 1.5845,
      "step": 3492
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4976675305442429e-05,
      "loss": 1.3217,
      "step": 3493
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4975194372454646e-05,
      "loss": 1.5389,
      "step": 3494
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4973713439466866e-05,
      "loss": 1.2404,
      "step": 3495
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4972232506479083e-05,
      "loss": 1.2929,
      "step": 3496
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.49707515734913e-05,
      "loss": 1.2469,
      "step": 3497
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.496927064050352e-05,
      "loss": 1.5178,
      "step": 3498
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4967789707515736e-05,
      "loss": 1.5129,
      "step": 3499
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4966308774527953e-05,
      "loss": 1.5109,
      "step": 3500
    },
    {
      "epoch": 0.77,
      "eval_loss": NaN,
      "eval_runtime": 4.3699,
      "eval_samples_per_second": 353.322,
      "eval_steps_per_second": 44.165,
      "step": 3500
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.496482784154017e-05,
      "loss": 1.4135,
      "step": 3501
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.496334690855239e-05,
      "loss": 1.5954,
      "step": 3502
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4961865975564607e-05,
      "loss": 1.2965,
      "step": 3503
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4960385042576823e-05,
      "loss": 1.4738,
      "step": 3504
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4958904109589042e-05,
      "loss": 1.5472,
      "step": 3505
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.495742317660126e-05,
      "loss": 1.2469,
      "step": 3506
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4955942243613477e-05,
      "loss": 1.1976,
      "step": 3507
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4954461310625697e-05,
      "loss": 1.3121,
      "step": 3508
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4952980377637914e-05,
      "loss": 1.5214,
      "step": 3509
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.495149944465013e-05,
      "loss": 1.4144,
      "step": 3510
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4950018511662348e-05,
      "loss": 1.3991,
      "step": 3511
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4948537578674567e-05,
      "loss": 1.4799,
      "step": 3512
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4947056645686784e-05,
      "loss": 1.466,
      "step": 3513
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.4945575712699e-05,
      "loss": 1.4415,
      "step": 3514
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4944094779711221e-05,
      "loss": 1.3713,
      "step": 3515
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4942613846723437e-05,
      "loss": 1.7359,
      "step": 3516
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4941132913735654e-05,
      "loss": 1.5724,
      "step": 3517
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4939651980747874e-05,
      "loss": 1.5511,
      "step": 3518
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4938171047760091e-05,
      "loss": 1.3353,
      "step": 3519
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4936690114772307e-05,
      "loss": 1.1946,
      "step": 3520
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4935209181784524e-05,
      "loss": 1.3645,
      "step": 3521
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4933728248796744e-05,
      "loss": 1.2274,
      "step": 3522
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4932247315808961e-05,
      "loss": 1.3463,
      "step": 3523
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4930766382821179e-05,
      "loss": 1.594,
      "step": 3524
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4929285449833398e-05,
      "loss": 1.3855,
      "step": 3525
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4927804516845614e-05,
      "loss": 1.589,
      "step": 3526
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4926323583857831e-05,
      "loss": 1.5287,
      "step": 3527
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4924842650870049e-05,
      "loss": 1.5046,
      "step": 3528
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4923361717882268e-05,
      "loss": 1.2958,
      "step": 3529
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4921880784894485e-05,
      "loss": 1.5363,
      "step": 3530
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4920399851906701e-05,
      "loss": 1.5276,
      "step": 3531
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.491891891891892e-05,
      "loss": 1.5282,
      "step": 3532
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4917437985931138e-05,
      "loss": 1.486,
      "step": 3533
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4915957052943355e-05,
      "loss": 1.2785,
      "step": 3534
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4914476119955575e-05,
      "loss": 1.4847,
      "step": 3535
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4912995186967792e-05,
      "loss": 1.4718,
      "step": 3536
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4911514253980008e-05,
      "loss": 1.3825,
      "step": 3537
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4910033320992226e-05,
      "loss": 1.2685,
      "step": 3538
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4908552388004445e-05,
      "loss": 1.4453,
      "step": 3539
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4907071455016662e-05,
      "loss": 1.3991,
      "step": 3540
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4905590522028878e-05,
      "loss": 1.2756,
      "step": 3541
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4904109589041097e-05,
      "loss": 1.6257,
      "step": 3542
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4902628656053315e-05,
      "loss": 1.4399,
      "step": 3543
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4901147723065532e-05,
      "loss": 1.2825,
      "step": 3544
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.489966679007775e-05,
      "loss": 1.5993,
      "step": 3545
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4898185857089969e-05,
      "loss": 1.4626,
      "step": 3546
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4896704924102185e-05,
      "loss": 1.5153,
      "step": 3547
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4895223991114402e-05,
      "loss": 1.4858,
      "step": 3548
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4893743058126622e-05,
      "loss": 1.2712,
      "step": 3549
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4892262125138839e-05,
      "loss": 1.3812,
      "step": 3550
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4890781192151057e-05,
      "loss": 1.6371,
      "step": 3551
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4889300259163276e-05,
      "loss": 1.5292,
      "step": 3552
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4887819326175492e-05,
      "loss": 1.2661,
      "step": 3553
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.488633839318771e-05,
      "loss": 1.31,
      "step": 3554
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4884857460199927e-05,
      "loss": 1.3332,
      "step": 3555
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4883376527212146e-05,
      "loss": 1.3125,
      "step": 3556
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4881895594224362e-05,
      "loss": 1.5185,
      "step": 3557
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.488041466123658e-05,
      "loss": 1.4632,
      "step": 3558
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.4878933728248798e-05,
      "loss": 1.3954,
      "step": 3559
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4877452795261016e-05,
      "loss": 1.3389,
      "step": 3560
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4875971862273234e-05,
      "loss": 1.4733,
      "step": 3561
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.487449092928545e-05,
      "loss": 1.4065,
      "step": 3562
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4873009996297669e-05,
      "loss": 1.3846,
      "step": 3563
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4871529063309886e-05,
      "loss": 1.2233,
      "step": 3564
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4870048130322104e-05,
      "loss": 1.4595,
      "step": 3565
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4868567197334323e-05,
      "loss": 1.3975,
      "step": 3566
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.486708626434654e-05,
      "loss": 1.8443,
      "step": 3567
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4865605331358756e-05,
      "loss": 1.5442,
      "step": 3568
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4864124398370975e-05,
      "loss": 1.4411,
      "step": 3569
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4862643465383193e-05,
      "loss": 1.4144,
      "step": 3570
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.486116253239541e-05,
      "loss": 1.4371,
      "step": 3571
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4859681599407628e-05,
      "loss": 1.3276,
      "step": 3572
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4858200666419847e-05,
      "loss": 1.8609,
      "step": 3573
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4856719733432063e-05,
      "loss": 1.2312,
      "step": 3574
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.485523880044428e-05,
      "loss": 1.2787,
      "step": 3575
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.48537578674565e-05,
      "loss": 1.4849,
      "step": 3576
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4852276934468717e-05,
      "loss": 1.3106,
      "step": 3577
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4850796001480933e-05,
      "loss": 1.5051,
      "step": 3578
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.484931506849315e-05,
      "loss": 1.2968,
      "step": 3579
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.484783413550537e-05,
      "loss": 1.3545,
      "step": 3580
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4846353202517587e-05,
      "loss": 1.4095,
      "step": 3581
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4844872269529805e-05,
      "loss": 1.5317,
      "step": 3582
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4843391336542024e-05,
      "loss": 1.5315,
      "step": 3583
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.484191040355424e-05,
      "loss": 1.3817,
      "step": 3584
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4840429470566457e-05,
      "loss": 1.528,
      "step": 3585
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4838948537578677e-05,
      "loss": 1.2617,
      "step": 3586
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4837467604590894e-05,
      "loss": 1.6862,
      "step": 3587
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4835986671603112e-05,
      "loss": 1.4315,
      "step": 3588
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4834505738615327e-05,
      "loss": 1.4867,
      "step": 3589
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4833024805627547e-05,
      "loss": 1.7012,
      "step": 3590
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4831543872639764e-05,
      "loss": 1.5083,
      "step": 3591
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4830062939651982e-05,
      "loss": 1.335,
      "step": 3592
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.48285820066642e-05,
      "loss": 1.3635,
      "step": 3593
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4827101073676418e-05,
      "loss": 1.2977,
      "step": 3594
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4825620140688634e-05,
      "loss": 1.4089,
      "step": 3595
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4824139207700852e-05,
      "loss": 1.4068,
      "step": 3596
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4822658274713071e-05,
      "loss": 1.398,
      "step": 3597
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4821177341725288e-05,
      "loss": 1.4098,
      "step": 3598
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4819696408737504e-05,
      "loss": 1.3962,
      "step": 3599
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4818215475749725e-05,
      "loss": 1.5618,
      "step": 3600
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4816734542761941e-05,
      "loss": 1.276,
      "step": 3601
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4815253609774158e-05,
      "loss": 1.5608,
      "step": 3602
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4813772676786378e-05,
      "loss": 1.7132,
      "step": 3603
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4812291743798595e-05,
      "loss": 1.3067,
      "step": 3604
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4810810810810811e-05,
      "loss": 1.565,
      "step": 3605
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4809329877823029e-05,
      "loss": 1.2173,
      "step": 3606
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4807848944835248e-05,
      "loss": 1.3511,
      "step": 3607
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4806368011847465e-05,
      "loss": 1.3836,
      "step": 3608
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4804887078859683e-05,
      "loss": 1.3121,
      "step": 3609
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4803406145871902e-05,
      "loss": 1.4927,
      "step": 3610
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4801925212884118e-05,
      "loss": 1.3705,
      "step": 3611
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4800444279896335e-05,
      "loss": 1.3679,
      "step": 3612
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4798963346908553e-05,
      "loss": 1.4502,
      "step": 3613
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4797482413920772e-05,
      "loss": 1.2713,
      "step": 3614
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.479600148093299e-05,
      "loss": 1.3962,
      "step": 3615
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4794520547945205e-05,
      "loss": 1.3229,
      "step": 3616
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4793039614957425e-05,
      "loss": 1.5733,
      "step": 3617
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4791558681969642e-05,
      "loss": 1.444,
      "step": 3618
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.479007774898186e-05,
      "loss": 1.5253,
      "step": 3619
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4788596815994079e-05,
      "loss": 1.2617,
      "step": 3620
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4787115883006295e-05,
      "loss": 1.3311,
      "step": 3621
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4785634950018512e-05,
      "loss": 1.4095,
      "step": 3622
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.478415401703073e-05,
      "loss": 1.4361,
      "step": 3623
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4782673084042949e-05,
      "loss": 1.4907,
      "step": 3624
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4781192151055166e-05,
      "loss": 1.3694,
      "step": 3625
    },
    {
      "epoch": 0.8,
      "eval_loss": NaN,
      "eval_runtime": 4.3747,
      "eval_samples_per_second": 352.939,
      "eval_steps_per_second": 44.117,
      "step": 3625
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4779711218067382e-05,
      "loss": 1.3062,
      "step": 3626
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4778230285079601e-05,
      "loss": 1.3361,
      "step": 3627
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4776749352091819e-05,
      "loss": 1.6124,
      "step": 3628
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4775268419104037e-05,
      "loss": 1.4913,
      "step": 3629
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4773787486116254e-05,
      "loss": 1.3833,
      "step": 3630
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4772306553128473e-05,
      "loss": 1.3589,
      "step": 3631
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4770825620140689e-05,
      "loss": 1.2601,
      "step": 3632
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4769344687152907e-05,
      "loss": 1.3462,
      "step": 3633
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4767863754165126e-05,
      "loss": 1.3386,
      "step": 3634
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4766382821177343e-05,
      "loss": 1.4011,
      "step": 3635
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.476490188818956e-05,
      "loss": 1.4649,
      "step": 3636
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.476342095520178e-05,
      "loss": 1.4196,
      "step": 3637
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4761940022213996e-05,
      "loss": 1.2861,
      "step": 3638
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4760459089226213e-05,
      "loss": 1.4908,
      "step": 3639
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4758978156238431e-05,
      "loss": 1.6613,
      "step": 3640
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.475749722325065e-05,
      "loss": 1.4061,
      "step": 3641
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4756016290262866e-05,
      "loss": 1.341,
      "step": 3642
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4754535357275083e-05,
      "loss": 1.3518,
      "step": 3643
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4753054424287303e-05,
      "loss": 1.4162,
      "step": 3644
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.475157349129952e-05,
      "loss": 1.5176,
      "step": 3645
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4750092558311738e-05,
      "loss": 1.2464,
      "step": 3646
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4748611625323957e-05,
      "loss": 1.6017,
      "step": 3647
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4747130692336173e-05,
      "loss": 1.5194,
      "step": 3648
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.474564975934839e-05,
      "loss": 1.3332,
      "step": 3649
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.4744168826360608e-05,
      "loss": 1.3931,
      "step": 3650
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4742687893372827e-05,
      "loss": 1.4077,
      "step": 3651
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4741206960385044e-05,
      "loss": 1.3251,
      "step": 3652
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.473972602739726e-05,
      "loss": 1.4526,
      "step": 3653
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.473824509440948e-05,
      "loss": 1.2553,
      "step": 3654
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4736764161421697e-05,
      "loss": 1.551,
      "step": 3655
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4735283228433915e-05,
      "loss": 1.4099,
      "step": 3656
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4733802295446132e-05,
      "loss": 1.3104,
      "step": 3657
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4732321362458351e-05,
      "loss": 1.4028,
      "step": 3658
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4730840429470567e-05,
      "loss": 1.3567,
      "step": 3659
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4729359496482785e-05,
      "loss": 1.2893,
      "step": 3660
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4727878563495004e-05,
      "loss": 1.5044,
      "step": 3661
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4726397630507221e-05,
      "loss": 1.705,
      "step": 3662
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4724916697519437e-05,
      "loss": 1.441,
      "step": 3663
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4723435764531658e-05,
      "loss": 1.4699,
      "step": 3664
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4721954831543874e-05,
      "loss": 1.3736,
      "step": 3665
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4720473898556091e-05,
      "loss": 1.2476,
      "step": 3666
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4718992965568309e-05,
      "loss": 1.1932,
      "step": 3667
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4717512032580528e-05,
      "loss": 1.4162,
      "step": 3668
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4716031099592744e-05,
      "loss": 1.572,
      "step": 3669
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4714550166604961e-05,
      "loss": 1.505,
      "step": 3670
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.471306923361718e-05,
      "loss": 1.3365,
      "step": 3671
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4711588300629398e-05,
      "loss": 1.295,
      "step": 3672
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4710107367641616e-05,
      "loss": 1.4211,
      "step": 3673
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4708626434653832e-05,
      "loss": 1.4137,
      "step": 3674
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.470714550166605e-05,
      "loss": 1.3312,
      "step": 3675
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4705664568678268e-05,
      "loss": 1.4064,
      "step": 3676
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4704183635690486e-05,
      "loss": 1.4419,
      "step": 3677
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4702702702702705e-05,
      "loss": 1.5818,
      "step": 3678
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4701221769714923e-05,
      "loss": 1.6298,
      "step": 3679
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4699740836727138e-05,
      "loss": 1.6178,
      "step": 3680
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4698259903739358e-05,
      "loss": 1.3929,
      "step": 3681
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4696778970751575e-05,
      "loss": 1.5139,
      "step": 3682
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4695298037763793e-05,
      "loss": 1.4647,
      "step": 3683
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4693817104776008e-05,
      "loss": 1.3475,
      "step": 3684
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4692336171788228e-05,
      "loss": 1.551,
      "step": 3685
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4690855238800445e-05,
      "loss": 1.5072,
      "step": 3686
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4689374305812663e-05,
      "loss": 1.2498,
      "step": 3687
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4687893372824882e-05,
      "loss": 1.1952,
      "step": 3688
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.46864124398371e-05,
      "loss": 1.4768,
      "step": 3689
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4684931506849315e-05,
      "loss": 1.3443,
      "step": 3690
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4683450573861533e-05,
      "loss": 1.4245,
      "step": 3691
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4681969640873752e-05,
      "loss": 1.622,
      "step": 3692
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.468048870788597e-05,
      "loss": 1.2504,
      "step": 3693
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4679007774898187e-05,
      "loss": 1.4966,
      "step": 3694
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4677526841910406e-05,
      "loss": 1.4336,
      "step": 3695
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.4676045908922622e-05,
      "loss": 1.4657,
      "step": 3696
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.467456497593484e-05,
      "loss": 1.4748,
      "step": 3697
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4673084042947059e-05,
      "loss": 1.4283,
      "step": 3698
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4671603109959276e-05,
      "loss": 1.3248,
      "step": 3699
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4670122176971494e-05,
      "loss": 1.736,
      "step": 3700
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.466864124398371e-05,
      "loss": 1.5128,
      "step": 3701
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4667160310995929e-05,
      "loss": 1.5983,
      "step": 3702
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4665679378008146e-05,
      "loss": 1.4244,
      "step": 3703
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4664198445020364e-05,
      "loss": 1.3737,
      "step": 3704
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4662717512032583e-05,
      "loss": 1.3865,
      "step": 3705
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4661236579044799e-05,
      "loss": 1.4968,
      "step": 3706
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4659755646057016e-05,
      "loss": 1.5394,
      "step": 3707
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4658274713069234e-05,
      "loss": 1.465,
      "step": 3708
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4656793780081453e-05,
      "loss": 1.4483,
      "step": 3709
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.465531284709367e-05,
      "loss": 1.5769,
      "step": 3710
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4653831914105886e-05,
      "loss": 1.5417,
      "step": 3711
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4652350981118106e-05,
      "loss": 1.2897,
      "step": 3712
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4650870048130323e-05,
      "loss": 1.4573,
      "step": 3713
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.464938911514254e-05,
      "loss": 1.3077,
      "step": 3714
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.464790818215476e-05,
      "loss": 1.2887,
      "step": 3715
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4646427249166977e-05,
      "loss": 1.2625,
      "step": 3716
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4644946316179193e-05,
      "loss": 1.3341,
      "step": 3717
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.464346538319141e-05,
      "loss": 1.3846,
      "step": 3718
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.464198445020363e-05,
      "loss": 1.4162,
      "step": 3719
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4640503517215847e-05,
      "loss": 1.5615,
      "step": 3720
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4639022584228065e-05,
      "loss": 1.4357,
      "step": 3721
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4637541651240284e-05,
      "loss": 1.3935,
      "step": 3722
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.46360607182525e-05,
      "loss": 1.4187,
      "step": 3723
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4634579785264718e-05,
      "loss": 1.5195,
      "step": 3724
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4633098852276935e-05,
      "loss": 1.4011,
      "step": 3725
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4631617919289154e-05,
      "loss": 1.2655,
      "step": 3726
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.463013698630137e-05,
      "loss": 1.2797,
      "step": 3727
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4628656053313588e-05,
      "loss": 1.5925,
      "step": 3728
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4627175120325807e-05,
      "loss": 1.5661,
      "step": 3729
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4625694187338024e-05,
      "loss": 1.4634,
      "step": 3730
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4624213254350242e-05,
      "loss": 1.553,
      "step": 3731
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4622732321362461e-05,
      "loss": 1.4634,
      "step": 3732
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4621251388374677e-05,
      "loss": 1.5092,
      "step": 3733
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4619770455386894e-05,
      "loss": 1.3823,
      "step": 3734
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4618289522399112e-05,
      "loss": 1.147,
      "step": 3735
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4616808589411331e-05,
      "loss": 1.2409,
      "step": 3736
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4615327656423549e-05,
      "loss": 1.5274,
      "step": 3737
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4613846723435764e-05,
      "loss": 1.319,
      "step": 3738
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4612365790447984e-05,
      "loss": 1.6254,
      "step": 3739
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4610884857460201e-05,
      "loss": 1.5435,
      "step": 3740
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.4609403924472419e-05,
      "loss": 1.8657,
      "step": 3741
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4607922991484636e-05,
      "loss": 1.3455,
      "step": 3742
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4606442058496855e-05,
      "loss": 1.2508,
      "step": 3743
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4604961125509071e-05,
      "loss": 1.5176,
      "step": 3744
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4603480192521289e-05,
      "loss": 1.5993,
      "step": 3745
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4601999259533508e-05,
      "loss": 1.4215,
      "step": 3746
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4600518326545726e-05,
      "loss": 1.5702,
      "step": 3747
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4599037393557941e-05,
      "loss": 1.553,
      "step": 3748
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.459755646057016e-05,
      "loss": 1.2161,
      "step": 3749
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4596075527582378e-05,
      "loss": 1.4469,
      "step": 3750
    },
    {
      "epoch": 0.83,
      "eval_loss": NaN,
      "eval_runtime": 4.3705,
      "eval_samples_per_second": 353.281,
      "eval_steps_per_second": 44.16,
      "step": 3750
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4594594594594596e-05,
      "loss": 1.331,
      "step": 3751
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4593113661606813e-05,
      "loss": 1.4394,
      "step": 3752
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4591632728619032e-05,
      "loss": 1.35,
      "step": 3753
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4590151795631248e-05,
      "loss": 1.2512,
      "step": 3754
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4588670862643466e-05,
      "loss": 1.4857,
      "step": 3755
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4587189929655685e-05,
      "loss": 1.5619,
      "step": 3756
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4585708996667902e-05,
      "loss": 1.3514,
      "step": 3757
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.458422806368012e-05,
      "loss": 1.2084,
      "step": 3758
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4582747130692336e-05,
      "loss": 1.3035,
      "step": 3759
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4581266197704555e-05,
      "loss": 1.6574,
      "step": 3760
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4579785264716772e-05,
      "loss": 1.2259,
      "step": 3761
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.457830433172899e-05,
      "loss": 1.903,
      "step": 3762
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.457682339874121e-05,
      "loss": 1.3832,
      "step": 3763
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4575342465753427e-05,
      "loss": 1.6323,
      "step": 3764
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4573861532765643e-05,
      "loss": 1.3596,
      "step": 3765
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4572380599777862e-05,
      "loss": 1.5978,
      "step": 3766
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.457089966679008e-05,
      "loss": 1.5645,
      "step": 3767
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4569418733802297e-05,
      "loss": 1.2736,
      "step": 3768
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4567937800814513e-05,
      "loss": 1.4048,
      "step": 3769
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4566456867826732e-05,
      "loss": 1.5922,
      "step": 3770
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.456497593483895e-05,
      "loss": 1.4253,
      "step": 3771
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4563495001851167e-05,
      "loss": 1.302,
      "step": 3772
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4562014068863386e-05,
      "loss": 1.6253,
      "step": 3773
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4560533135875604e-05,
      "loss": 1.3754,
      "step": 3774
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.455905220288782e-05,
      "loss": 1.3344,
      "step": 3775
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4557571269900037e-05,
      "loss": 1.4711,
      "step": 3776
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4556090336912256e-05,
      "loss": 1.4128,
      "step": 3777
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4554609403924474e-05,
      "loss": 1.3421,
      "step": 3778
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4553128470936691e-05,
      "loss": 1.3908,
      "step": 3779
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.455164753794891e-05,
      "loss": 1.5118,
      "step": 3780
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4550166604961126e-05,
      "loss": 1.6197,
      "step": 3781
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4548685671973344e-05,
      "loss": 1.3956,
      "step": 3782
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4547204738985563e-05,
      "loss": 1.5057,
      "step": 3783
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.454572380599778e-05,
      "loss": 1.3817,
      "step": 3784
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4544242873009998e-05,
      "loss": 1.4616,
      "step": 3785
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.4542761940022214e-05,
      "loss": 1.2429,
      "step": 3786
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4541281007034433e-05,
      "loss": 1.5703,
      "step": 3787
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.453980007404665e-05,
      "loss": 1.3392,
      "step": 3788
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4538319141058868e-05,
      "loss": 1.4365,
      "step": 3789
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4536838208071087e-05,
      "loss": 1.3819,
      "step": 3790
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4535357275083303e-05,
      "loss": 1.508,
      "step": 3791
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.453387634209552e-05,
      "loss": 1.5965,
      "step": 3792
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.453239540910774e-05,
      "loss": 1.4121,
      "step": 3793
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4530914476119957e-05,
      "loss": 1.2001,
      "step": 3794
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4529433543132175e-05,
      "loss": 1.4756,
      "step": 3795
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.452795261014439e-05,
      "loss": 1.4324,
      "step": 3796
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.452647167715661e-05,
      "loss": 1.2156,
      "step": 3797
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4524990744168827e-05,
      "loss": 1.6454,
      "step": 3798
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4523509811181045e-05,
      "loss": 1.6236,
      "step": 3799
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4522028878193264e-05,
      "loss": 1.4526,
      "step": 3800
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4520547945205482e-05,
      "loss": 1.4718,
      "step": 3801
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4519067012217697e-05,
      "loss": 1.5842,
      "step": 3802
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4517586079229915e-05,
      "loss": 1.64,
      "step": 3803
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4516105146242134e-05,
      "loss": 1.3816,
      "step": 3804
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4514624213254352e-05,
      "loss": 1.4906,
      "step": 3805
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.451314328026657e-05,
      "loss": 1.4484,
      "step": 3806
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4511662347278788e-05,
      "loss": 1.2768,
      "step": 3807
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4510181414291004e-05,
      "loss": 1.2632,
      "step": 3808
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4508700481303222e-05,
      "loss": 1.4287,
      "step": 3809
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4507219548315441e-05,
      "loss": 1.443,
      "step": 3810
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4505738615327658e-05,
      "loss": 1.6417,
      "step": 3811
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4504257682339874e-05,
      "loss": 1.5046,
      "step": 3812
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4502776749352092e-05,
      "loss": 1.3019,
      "step": 3813
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4501295816364311e-05,
      "loss": 1.3742,
      "step": 3814
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4499814883376529e-05,
      "loss": 1.3109,
      "step": 3815
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4498333950388746e-05,
      "loss": 1.5615,
      "step": 3816
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4496853017400965e-05,
      "loss": 1.418,
      "step": 3817
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4495372084413181e-05,
      "loss": 1.5397,
      "step": 3818
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4493891151425399e-05,
      "loss": 1.4595,
      "step": 3819
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4492410218437616e-05,
      "loss": 1.4304,
      "step": 3820
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4490929285449835e-05,
      "loss": 1.6016,
      "step": 3821
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4489448352462053e-05,
      "loss": 1.3082,
      "step": 3822
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4487967419474269e-05,
      "loss": 1.6886,
      "step": 3823
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4486486486486488e-05,
      "loss": 1.6617,
      "step": 3824
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4485005553498705e-05,
      "loss": 1.3062,
      "step": 3825
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4483524620510923e-05,
      "loss": 1.3101,
      "step": 3826
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4482043687523142e-05,
      "loss": 1.4999,
      "step": 3827
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.448056275453536e-05,
      "loss": 1.2638,
      "step": 3828
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4479081821547575e-05,
      "loss": 1.2502,
      "step": 3829
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4477600888559793e-05,
      "loss": 1.5701,
      "step": 3830
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.4476119955572012e-05,
      "loss": 1.2412,
      "step": 3831
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.447463902258423e-05,
      "loss": 1.341,
      "step": 3832
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4473158089596446e-05,
      "loss": 1.5864,
      "step": 3833
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4471677156608665e-05,
      "loss": 1.3023,
      "step": 3834
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4470196223620882e-05,
      "loss": 1.3379,
      "step": 3835
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.44687152906331e-05,
      "loss": 1.5749,
      "step": 3836
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4467234357645317e-05,
      "loss": 1.1928,
      "step": 3837
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4465753424657537e-05,
      "loss": 1.4631,
      "step": 3838
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4464272491669752e-05,
      "loss": 1.5074,
      "step": 3839
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.446279155868197e-05,
      "loss": 1.4496,
      "step": 3840
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4461310625694189e-05,
      "loss": 1.2872,
      "step": 3841
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4459829692706407e-05,
      "loss": 1.4294,
      "step": 3842
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4458348759718624e-05,
      "loss": 1.2595,
      "step": 3843
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4456867826730843e-05,
      "loss": 1.2752,
      "step": 3844
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4455386893743059e-05,
      "loss": 1.6616,
      "step": 3845
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4453905960755277e-05,
      "loss": 1.515,
      "step": 3846
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4452425027767494e-05,
      "loss": 1.4975,
      "step": 3847
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4450944094779713e-05,
      "loss": 1.4915,
      "step": 3848
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4449463161791931e-05,
      "loss": 1.3952,
      "step": 3849
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4447982228804147e-05,
      "loss": 1.334,
      "step": 3850
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4446501295816366e-05,
      "loss": 1.36,
      "step": 3851
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4445020362828583e-05,
      "loss": 1.4671,
      "step": 3852
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4443539429840801e-05,
      "loss": 1.1974,
      "step": 3853
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4442058496853017e-05,
      "loss": 1.5422,
      "step": 3854
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4440577563865236e-05,
      "loss": 1.4064,
      "step": 3855
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4439096630877454e-05,
      "loss": 1.4722,
      "step": 3856
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4437615697889671e-05,
      "loss": 1.6218,
      "step": 3857
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.443613476490189e-05,
      "loss": 1.3369,
      "step": 3858
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4434653831914108e-05,
      "loss": 1.3546,
      "step": 3859
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4433172898926324e-05,
      "loss": 1.7177,
      "step": 3860
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4431691965938543e-05,
      "loss": 1.3079,
      "step": 3861
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.443021103295076e-05,
      "loss": 1.3009,
      "step": 3862
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4428730099962978e-05,
      "loss": 1.4804,
      "step": 3863
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4427249166975195e-05,
      "loss": 1.3565,
      "step": 3864
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4425768233987415e-05,
      "loss": 1.6885,
      "step": 3865
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.442428730099963e-05,
      "loss": 1.6216,
      "step": 3866
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4422806368011848e-05,
      "loss": 1.4791,
      "step": 3867
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4421325435024067e-05,
      "loss": 1.3027,
      "step": 3868
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4419844502036285e-05,
      "loss": 1.5092,
      "step": 3869
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4418363569048502e-05,
      "loss": 1.5022,
      "step": 3870
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4416882636060718e-05,
      "loss": 1.3928,
      "step": 3871
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4415401703072937e-05,
      "loss": 1.5372,
      "step": 3872
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4413920770085155e-05,
      "loss": 1.4301,
      "step": 3873
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4412439837097372e-05,
      "loss": 1.161,
      "step": 3874
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4410958904109591e-05,
      "loss": 1.2277,
      "step": 3875
    },
    {
      "epoch": 0.85,
      "eval_loss": NaN,
      "eval_runtime": 4.382,
      "eval_samples_per_second": 352.354,
      "eval_steps_per_second": 44.044,
      "step": 3875
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4409477971121807e-05,
      "loss": 1.3654,
      "step": 3876
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.4407997038134025e-05,
      "loss": 1.4023,
      "step": 3877
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4406516105146244e-05,
      "loss": 1.305,
      "step": 3878
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4405035172158461e-05,
      "loss": 1.3245,
      "step": 3879
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4403554239170679e-05,
      "loss": 1.5574,
      "step": 3880
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4402073306182895e-05,
      "loss": 1.499,
      "step": 3881
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4400592373195114e-05,
      "loss": 1.2685,
      "step": 3882
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4399111440207332e-05,
      "loss": 1.602,
      "step": 3883
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4397630507219549e-05,
      "loss": 1.4554,
      "step": 3884
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4396149574231768e-05,
      "loss": 1.5269,
      "step": 3885
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4394668641243986e-05,
      "loss": 1.3571,
      "step": 3886
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4393187708256202e-05,
      "loss": 1.2452,
      "step": 3887
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4391706775268419e-05,
      "loss": 1.3759,
      "step": 3888
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4390225842280638e-05,
      "loss": 1.7329,
      "step": 3889
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4388744909292856e-05,
      "loss": 1.2907,
      "step": 3890
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4387263976305073e-05,
      "loss": 1.711,
      "step": 3891
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4385783043317293e-05,
      "loss": 1.4326,
      "step": 3892
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4384302110329508e-05,
      "loss": 1.5196,
      "step": 3893
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4382821177341726e-05,
      "loss": 1.7531,
      "step": 3894
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4381340244353945e-05,
      "loss": 1.4333,
      "step": 3895
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4379859311366163e-05,
      "loss": 1.5871,
      "step": 3896
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4378378378378378e-05,
      "loss": 1.5273,
      "step": 3897
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4376897445390596e-05,
      "loss": 1.5455,
      "step": 3898
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4375416512402815e-05,
      "loss": 1.5536,
      "step": 3899
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4373935579415033e-05,
      "loss": 1.202,
      "step": 3900
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.437245464642725e-05,
      "loss": 1.3837,
      "step": 3901
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.437097371343947e-05,
      "loss": 1.4608,
      "step": 3902
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4369492780451685e-05,
      "loss": 1.221,
      "step": 3903
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4368011847463903e-05,
      "loss": 1.3952,
      "step": 3904
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.436653091447612e-05,
      "loss": 1.5307,
      "step": 3905
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.436504998148834e-05,
      "loss": 1.504,
      "step": 3906
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4363569048500557e-05,
      "loss": 1.933,
      "step": 3907
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4362088115512773e-05,
      "loss": 1.6109,
      "step": 3908
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4360607182524992e-05,
      "loss": 1.2776,
      "step": 3909
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.435912624953721e-05,
      "loss": 1.3344,
      "step": 3910
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4357645316549427e-05,
      "loss": 1.4178,
      "step": 3911
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4356164383561646e-05,
      "loss": 1.2794,
      "step": 3912
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4354683450573864e-05,
      "loss": 1.3913,
      "step": 3913
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.435320251758608e-05,
      "loss": 1.5055,
      "step": 3914
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4351721584598297e-05,
      "loss": 1.3484,
      "step": 3915
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4350240651610516e-05,
      "loss": 1.4967,
      "step": 3916
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4348759718622734e-05,
      "loss": 1.5518,
      "step": 3917
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.434727878563495e-05,
      "loss": 1.5411,
      "step": 3918
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4345797852647169e-05,
      "loss": 1.6257,
      "step": 3919
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4344316919659386e-05,
      "loss": 1.3697,
      "step": 3920
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4342835986671604e-05,
      "loss": 1.5422,
      "step": 3921
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.4341355053683823e-05,
      "loss": 1.4068,
      "step": 3922
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.433987412069604e-05,
      "loss": 1.2497,
      "step": 3923
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4338393187708257e-05,
      "loss": 1.4147,
      "step": 3924
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4336912254720474e-05,
      "loss": 1.4728,
      "step": 3925
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4335431321732693e-05,
      "loss": 1.3424,
      "step": 3926
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.433395038874491e-05,
      "loss": 1.3024,
      "step": 3927
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4332469455757128e-05,
      "loss": 1.356,
      "step": 3928
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4330988522769347e-05,
      "loss": 1.3918,
      "step": 3929
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4329507589781563e-05,
      "loss": 1.2923,
      "step": 3930
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.432802665679378e-05,
      "loss": 1.7066,
      "step": 3931
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4326545723805998e-05,
      "loss": 1.4439,
      "step": 3932
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4325064790818218e-05,
      "loss": 1.6495,
      "step": 3933
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4323583857830435e-05,
      "loss": 1.4756,
      "step": 3934
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4322102924842651e-05,
      "loss": 1.3466,
      "step": 3935
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.432062199185487e-05,
      "loss": 1.4095,
      "step": 3936
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4319141058867088e-05,
      "loss": 1.2034,
      "step": 3937
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4317660125879305e-05,
      "loss": 1.4669,
      "step": 3938
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4316179192891524e-05,
      "loss": 1.5084,
      "step": 3939
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.431469825990374e-05,
      "loss": 1.6265,
      "step": 3940
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4313217326915958e-05,
      "loss": 1.24,
      "step": 3941
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4311736393928175e-05,
      "loss": 1.453,
      "step": 3942
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4310255460940394e-05,
      "loss": 1.5163,
      "step": 3943
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4308774527952612e-05,
      "loss": 1.3612,
      "step": 3944
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4307293594964828e-05,
      "loss": 1.4442,
      "step": 3945
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4305812661977047e-05,
      "loss": 1.3471,
      "step": 3946
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4304331728989264e-05,
      "loss": 1.3203,
      "step": 3947
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4302850796001482e-05,
      "loss": 1.21,
      "step": 3948
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.43013698630137e-05,
      "loss": 1.6403,
      "step": 3949
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4299888930025919e-05,
      "loss": 1.3345,
      "step": 3950
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4298407997038135e-05,
      "loss": 1.3886,
      "step": 3951
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4296927064050352e-05,
      "loss": 1.7588,
      "step": 3952
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4295446131062571e-05,
      "loss": 1.4638,
      "step": 3953
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4293965198074789e-05,
      "loss": 1.4549,
      "step": 3954
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4292484265087006e-05,
      "loss": 1.3429,
      "step": 3955
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4291003332099226e-05,
      "loss": 1.3276,
      "step": 3956
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4289522399111441e-05,
      "loss": 1.5283,
      "step": 3957
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4288041466123659e-05,
      "loss": 1.4221,
      "step": 3958
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4286560533135876e-05,
      "loss": 1.3673,
      "step": 3959
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4285079600148096e-05,
      "loss": 1.5597,
      "step": 3960
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4283598667160311e-05,
      "loss": 1.3391,
      "step": 3961
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4282117734172529e-05,
      "loss": 1.4267,
      "step": 3962
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4280636801184748e-05,
      "loss": 1.3096,
      "step": 3963
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4279155868196966e-05,
      "loss": 1.3275,
      "step": 3964
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4277674935209183e-05,
      "loss": 1.5613,
      "step": 3965
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4276194002221399e-05,
      "loss": 1.3901,
      "step": 3966
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4274713069233618e-05,
      "loss": 1.5226,
      "step": 3967
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.4273232136245836e-05,
      "loss": 1.3467,
      "step": 3968
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4271751203258053e-05,
      "loss": 1.7501,
      "step": 3969
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4270270270270272e-05,
      "loss": 1.3694,
      "step": 3970
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.426878933728249e-05,
      "loss": 1.3112,
      "step": 3971
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4267308404294706e-05,
      "loss": 1.4302,
      "step": 3972
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4265827471306925e-05,
      "loss": 1.4149,
      "step": 3973
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4264346538319143e-05,
      "loss": 1.5473,
      "step": 3974
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.426286560533136e-05,
      "loss": 1.3093,
      "step": 3975
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4261384672343576e-05,
      "loss": 1.3105,
      "step": 3976
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4259903739355797e-05,
      "loss": 1.4033,
      "step": 3977
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4258422806368013e-05,
      "loss": 1.832,
      "step": 3978
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.425694187338023e-05,
      "loss": 1.3671,
      "step": 3979
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.425546094039245e-05,
      "loss": 1.3453,
      "step": 3980
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4253980007404667e-05,
      "loss": 1.2065,
      "step": 3981
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4252499074416883e-05,
      "loss": 1.299,
      "step": 3982
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.42510181414291e-05,
      "loss": 1.4911,
      "step": 3983
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.424953720844132e-05,
      "loss": 1.4101,
      "step": 3984
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4248056275453537e-05,
      "loss": 1.2579,
      "step": 3985
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4246575342465754e-05,
      "loss": 1.6995,
      "step": 3986
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4245094409477974e-05,
      "loss": 1.4766,
      "step": 3987
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.424361347649019e-05,
      "loss": 1.592,
      "step": 3988
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4242132543502407e-05,
      "loss": 1.5408,
      "step": 3989
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4240651610514626e-05,
      "loss": 1.4633,
      "step": 3990
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4239170677526844e-05,
      "loss": 1.4415,
      "step": 3991
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4237689744539061e-05,
      "loss": 1.4986,
      "step": 3992
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4236208811551277e-05,
      "loss": 1.5972,
      "step": 3993
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4234727878563496e-05,
      "loss": 1.5285,
      "step": 3994
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4233246945575714e-05,
      "loss": 1.3398,
      "step": 3995
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4231766012587931e-05,
      "loss": 1.6906,
      "step": 3996
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.423028507960015e-05,
      "loss": 1.5649,
      "step": 3997
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4228804146612368e-05,
      "loss": 1.4121,
      "step": 3998
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4227323213624584e-05,
      "loss": 1.3705,
      "step": 3999
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4225842280636801e-05,
      "loss": 1.5621,
      "step": 4000
    },
    {
      "epoch": 0.88,
      "eval_loss": NaN,
      "eval_runtime": 4.3728,
      "eval_samples_per_second": 353.093,
      "eval_steps_per_second": 44.137,
      "step": 4000
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.422436134764902e-05,
      "loss": 1.38,
      "step": 4001
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4222880414661238e-05,
      "loss": 1.3819,
      "step": 4002
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4221399481673454e-05,
      "loss": 1.4955,
      "step": 4003
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4219918548685673e-05,
      "loss": 1.6022,
      "step": 4004
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.421843761569789e-05,
      "loss": 1.3107,
      "step": 4005
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4216956682710108e-05,
      "loss": 1.4311,
      "step": 4006
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4215475749722327e-05,
      "loss": 1.6004,
      "step": 4007
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4213994816734545e-05,
      "loss": 1.3379,
      "step": 4008
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.421251388374676e-05,
      "loss": 1.4361,
      "step": 4009
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4211032950758978e-05,
      "loss": 1.3902,
      "step": 4010
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4209552017771197e-05,
      "loss": 1.3583,
      "step": 4011
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4208071084783415e-05,
      "loss": 1.5042,
      "step": 4012
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4206590151795632e-05,
      "loss": 1.3662,
      "step": 4013
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4205109218807852e-05,
      "loss": 1.2743,
      "step": 4014
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4203628285820067e-05,
      "loss": 1.5094,
      "step": 4015
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4202147352832285e-05,
      "loss": 1.4337,
      "step": 4016
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4200666419844503e-05,
      "loss": 1.3389,
      "step": 4017
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4199185486856722e-05,
      "loss": 1.3535,
      "step": 4018
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.419770455386894e-05,
      "loss": 1.5236,
      "step": 4019
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4196223620881155e-05,
      "loss": 1.4799,
      "step": 4020
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4194742687893374e-05,
      "loss": 1.3227,
      "step": 4021
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4193261754905592e-05,
      "loss": 1.3075,
      "step": 4022
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.419178082191781e-05,
      "loss": 1.3839,
      "step": 4023
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4190299888930029e-05,
      "loss": 1.4328,
      "step": 4024
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4188818955942244e-05,
      "loss": 1.3571,
      "step": 4025
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4187338022954462e-05,
      "loss": 1.5316,
      "step": 4026
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.418585708996668e-05,
      "loss": 1.613,
      "step": 4027
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4184376156978899e-05,
      "loss": 1.6234,
      "step": 4028
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4182895223991116e-05,
      "loss": 1.4456,
      "step": 4029
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4181414291003332e-05,
      "loss": 1.1814,
      "step": 4030
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4179933358015551e-05,
      "loss": 1.6871,
      "step": 4031
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4178452425027769e-05,
      "loss": 1.4902,
      "step": 4032
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4176971492039986e-05,
      "loss": 1.185,
      "step": 4033
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4175490559052204e-05,
      "loss": 1.592,
      "step": 4034
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4174009626064423e-05,
      "loss": 1.4418,
      "step": 4035
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4172528693076639e-05,
      "loss": 1.4645,
      "step": 4036
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4171047760088856e-05,
      "loss": 1.6559,
      "step": 4037
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4169566827101075e-05,
      "loss": 1.4443,
      "step": 4038
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4168085894113293e-05,
      "loss": 1.7108,
      "step": 4039
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4166604961125509e-05,
      "loss": 1.5097,
      "step": 4040
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.416512402813773e-05,
      "loss": 1.3384,
      "step": 4041
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4163643095149946e-05,
      "loss": 1.5003,
      "step": 4042
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4162162162162163e-05,
      "loss": 1.4997,
      "step": 4043
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.416068122917438e-05,
      "loss": 1.5841,
      "step": 4044
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.41592002961866e-05,
      "loss": 1.1745,
      "step": 4045
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4157719363198816e-05,
      "loss": 1.4316,
      "step": 4046
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4156238430211033e-05,
      "loss": 1.4409,
      "step": 4047
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4154757497223252e-05,
      "loss": 1.4124,
      "step": 4048
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.415327656423547e-05,
      "loss": 1.1707,
      "step": 4049
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4151795631247687e-05,
      "loss": 1.4655,
      "step": 4050
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4150314698259907e-05,
      "loss": 1.3862,
      "step": 4051
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4148833765272122e-05,
      "loss": 1.4948,
      "step": 4052
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.414735283228434e-05,
      "loss": 1.3165,
      "step": 4053
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4145871899296557e-05,
      "loss": 1.5557,
      "step": 4054
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4144390966308777e-05,
      "loss": 1.377,
      "step": 4055
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4142910033320994e-05,
      "loss": 1.5515,
      "step": 4056
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.414142910033321e-05,
      "loss": 1.3194,
      "step": 4057
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.413994816734543e-05,
      "loss": 1.3915,
      "step": 4058
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4138467234357647e-05,
      "loss": 1.2528,
      "step": 4059
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4136986301369864e-05,
      "loss": 1.3889,
      "step": 4060
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.413550536838208e-05,
      "loss": 1.2083,
      "step": 4061
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4134024435394301e-05,
      "loss": 1.4152,
      "step": 4062
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4132543502406517e-05,
      "loss": 1.4186,
      "step": 4063
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4131062569418734e-05,
      "loss": 1.2268,
      "step": 4064
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4129581636430954e-05,
      "loss": 1.5967,
      "step": 4065
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4128100703443171e-05,
      "loss": 1.4316,
      "step": 4066
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4126619770455387e-05,
      "loss": 1.57,
      "step": 4067
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4125138837467606e-05,
      "loss": 1.5512,
      "step": 4068
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4123657904479824e-05,
      "loss": 1.492,
      "step": 4069
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4122176971492041e-05,
      "loss": 1.5103,
      "step": 4070
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4120696038504259e-05,
      "loss": 1.3677,
      "step": 4071
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4119215105516478e-05,
      "loss": 1.3027,
      "step": 4072
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4117734172528694e-05,
      "loss": 1.3644,
      "step": 4073
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4116253239540911e-05,
      "loss": 1.4177,
      "step": 4074
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.411477230655313e-05,
      "loss": 1.509,
      "step": 4075
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4113291373565348e-05,
      "loss": 1.3137,
      "step": 4076
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4111810440577565e-05,
      "loss": 1.2403,
      "step": 4077
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4110329507589781e-05,
      "loss": 1.4019,
      "step": 4078
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4108848574602e-05,
      "loss": 1.1851,
      "step": 4079
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4107367641614218e-05,
      "loss": 1.415,
      "step": 4080
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4105886708626435e-05,
      "loss": 1.4114,
      "step": 4081
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4104405775638655e-05,
      "loss": 1.3442,
      "step": 4082
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4102924842650872e-05,
      "loss": 1.6339,
      "step": 4083
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4101443909663088e-05,
      "loss": 1.3828,
      "step": 4084
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4099962976675307e-05,
      "loss": 1.3477,
      "step": 4085
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4098482043687525e-05,
      "loss": 1.5319,
      "step": 4086
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4097001110699742e-05,
      "loss": 1.5008,
      "step": 4087
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4095520177711958e-05,
      "loss": 1.4502,
      "step": 4088
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4094039244724177e-05,
      "loss": 1.3167,
      "step": 4089
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4092558311736395e-05,
      "loss": 1.2811,
      "step": 4090
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4091077378748612e-05,
      "loss": 1.4782,
      "step": 4091
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4089596445760832e-05,
      "loss": 1.2745,
      "step": 4092
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4088115512773049e-05,
      "loss": 1.4516,
      "step": 4093
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4086634579785265e-05,
      "loss": 1.4755,
      "step": 4094
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4085153646797482e-05,
      "loss": 1.3506,
      "step": 4095
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4083672713809702e-05,
      "loss": 1.369,
      "step": 4096
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4082191780821919e-05,
      "loss": 1.4631,
      "step": 4097
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4080710847834137e-05,
      "loss": 1.0673,
      "step": 4098
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4079229914846356e-05,
      "loss": 1.6404,
      "step": 4099
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4077748981858572e-05,
      "loss": 1.5815,
      "step": 4100
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.407626804887079e-05,
      "loss": 1.3164,
      "step": 4101
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4074787115883008e-05,
      "loss": 1.4,
      "step": 4102
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4073306182895226e-05,
      "loss": 1.3293,
      "step": 4103
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.4071825249907442e-05,
      "loss": 1.2298,
      "step": 4104
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.407034431691966e-05,
      "loss": 1.4137,
      "step": 4105
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4068863383931878e-05,
      "loss": 1.3667,
      "step": 4106
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4067382450944096e-05,
      "loss": 1.3922,
      "step": 4107
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4065901517956314e-05,
      "loss": 1.5713,
      "step": 4108
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4064420584968533e-05,
      "loss": 1.8738,
      "step": 4109
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4062939651980749e-05,
      "loss": 1.4309,
      "step": 4110
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4061458718992966e-05,
      "loss": 1.1215,
      "step": 4111
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4059977786005184e-05,
      "loss": 1.5406,
      "step": 4112
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4058496853017403e-05,
      "loss": 1.584,
      "step": 4113
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.405701592002962e-05,
      "loss": 1.3738,
      "step": 4114
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4055534987041836e-05,
      "loss": 1.3567,
      "step": 4115
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4054054054054055e-05,
      "loss": 1.3611,
      "step": 4116
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4052573121066273e-05,
      "loss": 1.6234,
      "step": 4117
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.405109218807849e-05,
      "loss": 1.7354,
      "step": 4118
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.404961125509071e-05,
      "loss": 1.2712,
      "step": 4119
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4048130322102927e-05,
      "loss": 1.1577,
      "step": 4120
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4046649389115143e-05,
      "loss": 1.5611,
      "step": 4121
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.404516845612736e-05,
      "loss": 1.3159,
      "step": 4122
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.404368752313958e-05,
      "loss": 1.4839,
      "step": 4123
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4042206590151797e-05,
      "loss": 1.5865,
      "step": 4124
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4040725657164013e-05,
      "loss": 1.2372,
      "step": 4125
    },
    {
      "epoch": 0.91,
      "eval_loss": NaN,
      "eval_runtime": 4.384,
      "eval_samples_per_second": 352.188,
      "eval_steps_per_second": 44.024,
      "step": 4125
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4039244724176234e-05,
      "loss": 1.3603,
      "step": 4126
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.403776379118845e-05,
      "loss": 1.347,
      "step": 4127
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4036282858200667e-05,
      "loss": 1.0635,
      "step": 4128
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4034801925212885e-05,
      "loss": 1.6206,
      "step": 4129
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4033320992225104e-05,
      "loss": 1.2906,
      "step": 4130
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.403184005923732e-05,
      "loss": 1.3382,
      "step": 4131
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4030359126249537e-05,
      "loss": 1.1371,
      "step": 4132
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4028878193261757e-05,
      "loss": 1.3947,
      "step": 4133
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4027397260273974e-05,
      "loss": 1.1863,
      "step": 4134
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4025916327286192e-05,
      "loss": 1.449,
      "step": 4135
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.402443539429841e-05,
      "loss": 1.3391,
      "step": 4136
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4022954461310627e-05,
      "loss": 1.5387,
      "step": 4137
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4021473528322844e-05,
      "loss": 1.3235,
      "step": 4138
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4019992595335062e-05,
      "loss": 1.4975,
      "step": 4139
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.401851166234728e-05,
      "loss": 1.5295,
      "step": 4140
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4017030729359498e-05,
      "loss": 1.1868,
      "step": 4141
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4015549796371714e-05,
      "loss": 1.2055,
      "step": 4142
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4014068863383933e-05,
      "loss": 1.2882,
      "step": 4143
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4012587930396151e-05,
      "loss": 1.4428,
      "step": 4144
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4011106997408368e-05,
      "loss": 1.4257,
      "step": 4145
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4009626064420584e-05,
      "loss": 1.4313,
      "step": 4146
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4008145131432805e-05,
      "loss": 1.2411,
      "step": 4147
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4006664198445021e-05,
      "loss": 1.715,
      "step": 4148
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.4005183265457238e-05,
      "loss": 1.4838,
      "step": 4149
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4003702332469458e-05,
      "loss": 1.2906,
      "step": 4150
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4002221399481675e-05,
      "loss": 1.2576,
      "step": 4151
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.4000740466493891e-05,
      "loss": 1.4675,
      "step": 4152
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.399925953350611e-05,
      "loss": 1.4628,
      "step": 4153
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3997778600518328e-05,
      "loss": 1.5884,
      "step": 4154
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3996297667530545e-05,
      "loss": 1.3441,
      "step": 4155
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3994816734542763e-05,
      "loss": 1.3474,
      "step": 4156
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3993335801554982e-05,
      "loss": 1.3141,
      "step": 4157
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3991854868567198e-05,
      "loss": 1.495,
      "step": 4158
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3990373935579415e-05,
      "loss": 1.516,
      "step": 4159
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3988893002591635e-05,
      "loss": 1.6827,
      "step": 4160
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3987412069603852e-05,
      "loss": 1.505,
      "step": 4161
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.398593113661607e-05,
      "loss": 1.5327,
      "step": 4162
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3984450203628285e-05,
      "loss": 1.6675,
      "step": 4163
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3982969270640505e-05,
      "loss": 1.5126,
      "step": 4164
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3981488337652722e-05,
      "loss": 1.4366,
      "step": 4165
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.398000740466494e-05,
      "loss": 1.3831,
      "step": 4166
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3978526471677159e-05,
      "loss": 1.3566,
      "step": 4167
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3977045538689375e-05,
      "loss": 1.4548,
      "step": 4168
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3975564605701592e-05,
      "loss": 1.3575,
      "step": 4169
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3974083672713811e-05,
      "loss": 1.3124,
      "step": 4170
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3972602739726029e-05,
      "loss": 1.2414,
      "step": 4171
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3971121806738246e-05,
      "loss": 1.2885,
      "step": 4172
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3969640873750462e-05,
      "loss": 1.6647,
      "step": 4173
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3968159940762681e-05,
      "loss": 1.4644,
      "step": 4174
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3966679007774899e-05,
      "loss": 1.2816,
      "step": 4175
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3965198074787117e-05,
      "loss": 1.3402,
      "step": 4176
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3963717141799336e-05,
      "loss": 1.3864,
      "step": 4177
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3962236208811553e-05,
      "loss": 1.3761,
      "step": 4178
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3960755275823769e-05,
      "loss": 1.4323,
      "step": 4179
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3959274342835988e-05,
      "loss": 1.3826,
      "step": 4180
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3957793409848206e-05,
      "loss": 1.5559,
      "step": 4181
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3956312476860423e-05,
      "loss": 1.4087,
      "step": 4182
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.395483154387264e-05,
      "loss": 1.3157,
      "step": 4183
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.395335061088486e-05,
      "loss": 1.3499,
      "step": 4184
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3951869677897076e-05,
      "loss": 1.3244,
      "step": 4185
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3950388744909293e-05,
      "loss": 1.3154,
      "step": 4186
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3948907811921513e-05,
      "loss": 1.5499,
      "step": 4187
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.394742687893373e-05,
      "loss": 1.3379,
      "step": 4188
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3945945945945946e-05,
      "loss": 1.41,
      "step": 4189
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3944465012958163e-05,
      "loss": 1.4052,
      "step": 4190
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3942984079970383e-05,
      "loss": 1.5176,
      "step": 4191
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.39415031469826e-05,
      "loss": 1.6387,
      "step": 4192
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3940022213994818e-05,
      "loss": 1.3318,
      "step": 4193
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.3938541281007037e-05,
      "loss": 1.323,
      "step": 4194
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3937060348019253e-05,
      "loss": 1.5991,
      "step": 4195
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.393557941503147e-05,
      "loss": 1.5193,
      "step": 4196
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.393409848204369e-05,
      "loss": 1.3123,
      "step": 4197
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3932617549055907e-05,
      "loss": 1.3403,
      "step": 4198
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3931136616068124e-05,
      "loss": 1.3397,
      "step": 4199
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.392965568308034e-05,
      "loss": 1.4019,
      "step": 4200
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.392817475009256e-05,
      "loss": 1.2319,
      "step": 4201
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3926693817104777e-05,
      "loss": 1.4268,
      "step": 4202
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3925212884116995e-05,
      "loss": 1.5414,
      "step": 4203
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3923731951129214e-05,
      "loss": 1.7303,
      "step": 4204
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3922251018141431e-05,
      "loss": 1.2429,
      "step": 4205
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3920770085153647e-05,
      "loss": 1.4269,
      "step": 4206
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3919289152165865e-05,
      "loss": 1.4031,
      "step": 4207
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3917808219178084e-05,
      "loss": 1.4941,
      "step": 4208
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3916327286190301e-05,
      "loss": 1.2568,
      "step": 4209
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3914846353202517e-05,
      "loss": 1.3606,
      "step": 4210
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3913365420214738e-05,
      "loss": 1.4833,
      "step": 4211
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3911884487226954e-05,
      "loss": 1.4667,
      "step": 4212
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3910403554239171e-05,
      "loss": 1.2447,
      "step": 4213
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.390892262125139e-05,
      "loss": 1.1673,
      "step": 4214
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3907441688263608e-05,
      "loss": 1.5909,
      "step": 4215
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3905960755275824e-05,
      "loss": 1.4246,
      "step": 4216
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3904479822288041e-05,
      "loss": 1.5854,
      "step": 4217
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.390299888930026e-05,
      "loss": 1.508,
      "step": 4218
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3901517956312478e-05,
      "loss": 1.4652,
      "step": 4219
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3900037023324696e-05,
      "loss": 1.4149,
      "step": 4220
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3898556090336915e-05,
      "loss": 1.2143,
      "step": 4221
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.389707515734913e-05,
      "loss": 1.4949,
      "step": 4222
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3895594224361348e-05,
      "loss": 1.5348,
      "step": 4223
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3894113291373566e-05,
      "loss": 1.4359,
      "step": 4224
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3892632358385785e-05,
      "loss": 1.4291,
      "step": 4225
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3891151425398003e-05,
      "loss": 1.547,
      "step": 4226
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3889670492410218e-05,
      "loss": 1.3752,
      "step": 4227
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3888189559422438e-05,
      "loss": 1.3149,
      "step": 4228
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3886708626434655e-05,
      "loss": 1.125,
      "step": 4229
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3885227693446873e-05,
      "loss": 1.5318,
      "step": 4230
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3883746760459092e-05,
      "loss": 1.3019,
      "step": 4231
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3882265827471308e-05,
      "loss": 1.4947,
      "step": 4232
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3880784894483525e-05,
      "loss": 1.2938,
      "step": 4233
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3879303961495743e-05,
      "loss": 1.5225,
      "step": 4234
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3877823028507962e-05,
      "loss": 1.344,
      "step": 4235
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.387634209552018e-05,
      "loss": 1.5179,
      "step": 4236
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3874861162532395e-05,
      "loss": 1.4573,
      "step": 4237
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3873380229544614e-05,
      "loss": 1.3901,
      "step": 4238
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.3871899296556832e-05,
      "loss": 1.4306,
      "step": 4239
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.387041836356905e-05,
      "loss": 1.5567,
      "step": 4240
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3868937430581267e-05,
      "loss": 1.5416,
      "step": 4241
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3867456497593486e-05,
      "loss": 1.4396,
      "step": 4242
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3865975564605702e-05,
      "loss": 1.47,
      "step": 4243
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.386449463161792e-05,
      "loss": 1.5811,
      "step": 4244
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3863013698630139e-05,
      "loss": 1.2677,
      "step": 4245
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3861532765642356e-05,
      "loss": 1.372,
      "step": 4246
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3860051832654574e-05,
      "loss": 1.291,
      "step": 4247
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3858570899666793e-05,
      "loss": 1.4125,
      "step": 4248
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3857089966679009e-05,
      "loss": 1.1841,
      "step": 4249
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3855609033691226e-05,
      "loss": 1.449,
      "step": 4250
    },
    {
      "epoch": 0.94,
      "eval_loss": NaN,
      "eval_runtime": 4.3715,
      "eval_samples_per_second": 353.2,
      "eval_steps_per_second": 44.15,
      "step": 4250
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3854128100703444e-05,
      "loss": 1.3972,
      "step": 4251
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3852647167715663e-05,
      "loss": 1.3189,
      "step": 4252
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3851166234727879e-05,
      "loss": 1.3195,
      "step": 4253
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3849685301740096e-05,
      "loss": 1.5291,
      "step": 4254
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3848204368752316e-05,
      "loss": 1.4335,
      "step": 4255
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3846723435764533e-05,
      "loss": 1.4415,
      "step": 4256
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.384524250277675e-05,
      "loss": 1.3667,
      "step": 4257
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3843761569788966e-05,
      "loss": 1.2132,
      "step": 4258
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3842280636801186e-05,
      "loss": 1.5133,
      "step": 4259
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3840799703813403e-05,
      "loss": 1.2439,
      "step": 4260
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.383931877082562e-05,
      "loss": 1.3664,
      "step": 4261
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.383783783783784e-05,
      "loss": 1.4478,
      "step": 4262
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3836356904850057e-05,
      "loss": 1.3876,
      "step": 4263
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3834875971862273e-05,
      "loss": 1.5901,
      "step": 4264
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3833395038874492e-05,
      "loss": 1.5251,
      "step": 4265
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.383191410588671e-05,
      "loss": 1.1781,
      "step": 4266
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3830433172898927e-05,
      "loss": 1.315,
      "step": 4267
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3828952239911145e-05,
      "loss": 1.334,
      "step": 4268
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3827471306923364e-05,
      "loss": 1.4092,
      "step": 4269
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.382599037393558e-05,
      "loss": 1.7247,
      "step": 4270
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3824509440947798e-05,
      "loss": 1.3429,
      "step": 4271
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3823028507960017e-05,
      "loss": 1.422,
      "step": 4272
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3821547574972234e-05,
      "loss": 1.5929,
      "step": 4273
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.382006664198445e-05,
      "loss": 1.5554,
      "step": 4274
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3818585708996668e-05,
      "loss": 1.2395,
      "step": 4275
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3817104776008887e-05,
      "loss": 1.5321,
      "step": 4276
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3815623843021104e-05,
      "loss": 1.594,
      "step": 4277
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3814142910033322e-05,
      "loss": 1.5464,
      "step": 4278
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3812661977045541e-05,
      "loss": 1.3562,
      "step": 4279
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3811181044057757e-05,
      "loss": 1.3533,
      "step": 4280
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3809700111069974e-05,
      "loss": 1.5756,
      "step": 4281
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3808219178082194e-05,
      "loss": 1.4889,
      "step": 4282
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3806738245094411e-05,
      "loss": 1.4279,
      "step": 4283
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3805257312106629e-05,
      "loss": 1.491,
      "step": 4284
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.3803776379118844e-05,
      "loss": 1.4623,
      "step": 4285
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3802295446131064e-05,
      "loss": 1.4332,
      "step": 4286
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3800814513143281e-05,
      "loss": 1.3187,
      "step": 4287
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3799333580155499e-05,
      "loss": 1.4028,
      "step": 4288
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3797852647167718e-05,
      "loss": 1.705,
      "step": 4289
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3796371714179935e-05,
      "loss": 1.5245,
      "step": 4290
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3794890781192151e-05,
      "loss": 1.6157,
      "step": 4291
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3793409848204369e-05,
      "loss": 1.3416,
      "step": 4292
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3791928915216588e-05,
      "loss": 1.2859,
      "step": 4293
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3790447982228806e-05,
      "loss": 1.4697,
      "step": 4294
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3788967049241021e-05,
      "loss": 1.4807,
      "step": 4295
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.378748611625324e-05,
      "loss": 1.3419,
      "step": 4296
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3786005183265458e-05,
      "loss": 1.4988,
      "step": 4297
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3784524250277676e-05,
      "loss": 1.3431,
      "step": 4298
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3783043317289895e-05,
      "loss": 1.2757,
      "step": 4299
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3781562384302112e-05,
      "loss": 1.5165,
      "step": 4300
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3780081451314328e-05,
      "loss": 1.3391,
      "step": 4301
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3778600518326546e-05,
      "loss": 1.2205,
      "step": 4302
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3777119585338765e-05,
      "loss": 1.5087,
      "step": 4303
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3775638652350982e-05,
      "loss": 1.4406,
      "step": 4304
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.37741577193632e-05,
      "loss": 1.5815,
      "step": 4305
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3772676786375419e-05,
      "loss": 1.3414,
      "step": 4306
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3771195853387635e-05,
      "loss": 1.5948,
      "step": 4307
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3769714920399852e-05,
      "loss": 1.2536,
      "step": 4308
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.376823398741207e-05,
      "loss": 1.3401,
      "step": 4309
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.376675305442429e-05,
      "loss": 1.6701,
      "step": 4310
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3765272121436507e-05,
      "loss": 1.5406,
      "step": 4311
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3763791188448723e-05,
      "loss": 1.3628,
      "step": 4312
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3762310255460942e-05,
      "loss": 1.2982,
      "step": 4313
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.376082932247316e-05,
      "loss": 1.4441,
      "step": 4314
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3759348389485377e-05,
      "loss": 1.4241,
      "step": 4315
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3757867456497596e-05,
      "loss": 1.4768,
      "step": 4316
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3756386523509812e-05,
      "loss": 1.4858,
      "step": 4317
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.375490559052203e-05,
      "loss": 1.4962,
      "step": 4318
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3753424657534247e-05,
      "loss": 1.3407,
      "step": 4319
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3751943724546466e-05,
      "loss": 1.3317,
      "step": 4320
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3750462791558684e-05,
      "loss": 1.4766,
      "step": 4321
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.37489818585709e-05,
      "loss": 1.45,
      "step": 4322
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3747500925583119e-05,
      "loss": 1.4864,
      "step": 4323
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3746019992595336e-05,
      "loss": 1.448,
      "step": 4324
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3744539059607554e-05,
      "loss": 1.3472,
      "step": 4325
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3743058126619773e-05,
      "loss": 1.553,
      "step": 4326
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.374157719363199e-05,
      "loss": 1.1852,
      "step": 4327
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3740096260644206e-05,
      "loss": 1.5549,
      "step": 4328
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3738615327656424e-05,
      "loss": 1.4068,
      "step": 4329
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.3737134394668643e-05,
      "loss": 1.41,
      "step": 4330
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.373565346168086e-05,
      "loss": 1.6293,
      "step": 4331
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3734172528693078e-05,
      "loss": 1.3409,
      "step": 4332
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3732691595705297e-05,
      "loss": 1.5542,
      "step": 4333
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3731210662717513e-05,
      "loss": 1.498,
      "step": 4334
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.372972972972973e-05,
      "loss": 1.4121,
      "step": 4335
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3728248796741948e-05,
      "loss": 1.2933,
      "step": 4336
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3726767863754167e-05,
      "loss": 1.3896,
      "step": 4337
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3725286930766383e-05,
      "loss": 1.2424,
      "step": 4338
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.37238059977786e-05,
      "loss": 1.4348,
      "step": 4339
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.372232506479082e-05,
      "loss": 1.2218,
      "step": 4340
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3720844131803037e-05,
      "loss": 1.3503,
      "step": 4341
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3719363198815255e-05,
      "loss": 1.5182,
      "step": 4342
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3717882265827474e-05,
      "loss": 1.3352,
      "step": 4343
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.371640133283969e-05,
      "loss": 1.4165,
      "step": 4344
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3714920399851907e-05,
      "loss": 1.495,
      "step": 4345
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3713439466864125e-05,
      "loss": 1.2077,
      "step": 4346
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3711958533876344e-05,
      "loss": 1.349,
      "step": 4347
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3710477600888562e-05,
      "loss": 1.3837,
      "step": 4348
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3708996667900777e-05,
      "loss": 1.4381,
      "step": 4349
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3707515734912997e-05,
      "loss": 1.338,
      "step": 4350
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3706034801925214e-05,
      "loss": 1.431,
      "step": 4351
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3704553868937432e-05,
      "loss": 1.4027,
      "step": 4352
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.370307293594965e-05,
      "loss": 1.2584,
      "step": 4353
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3701592002961868e-05,
      "loss": 1.3281,
      "step": 4354
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3700111069974084e-05,
      "loss": 1.4298,
      "step": 4355
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3698630136986302e-05,
      "loss": 1.234,
      "step": 4356
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3697149203998521e-05,
      "loss": 1.221,
      "step": 4357
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3695668271010738e-05,
      "loss": 1.3294,
      "step": 4358
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3694187338022954e-05,
      "loss": 1.4106,
      "step": 4359
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3692706405035174e-05,
      "loss": 1.4302,
      "step": 4360
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3691225472047391e-05,
      "loss": 1.4256,
      "step": 4361
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3689744539059609e-05,
      "loss": 1.61,
      "step": 4362
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3688263606071826e-05,
      "loss": 1.1962,
      "step": 4363
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3686782673084045e-05,
      "loss": 1.3089,
      "step": 4364
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3685301740096261e-05,
      "loss": 1.5239,
      "step": 4365
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3683820807108479e-05,
      "loss": 1.2507,
      "step": 4366
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3682339874120698e-05,
      "loss": 1.1975,
      "step": 4367
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3680858941132915e-05,
      "loss": 1.3447,
      "step": 4368
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3679378008145133e-05,
      "loss": 1.5191,
      "step": 4369
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3677897075157349e-05,
      "loss": 1.509,
      "step": 4370
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3676416142169568e-05,
      "loss": 1.4057,
      "step": 4371
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3674935209181785e-05,
      "loss": 1.2253,
      "step": 4372
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3673454276194003e-05,
      "loss": 1.2164,
      "step": 4373
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3671973343206222e-05,
      "loss": 1.3795,
      "step": 4374
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.367049241021844e-05,
      "loss": 1.4169,
      "step": 4375
    },
    {
      "epoch": 0.96,
      "eval_loss": NaN,
      "eval_runtime": 4.3763,
      "eval_samples_per_second": 352.807,
      "eval_steps_per_second": 44.101,
      "step": 4375
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.3669011477230655e-05,
      "loss": 1.2801,
      "step": 4376
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3667530544242875e-05,
      "loss": 1.4257,
      "step": 4377
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3666049611255092e-05,
      "loss": 1.3902,
      "step": 4378
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.366456867826731e-05,
      "loss": 1.4188,
      "step": 4379
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3663087745279526e-05,
      "loss": 1.4326,
      "step": 4380
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3661606812291745e-05,
      "loss": 1.5001,
      "step": 4381
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3660125879303962e-05,
      "loss": 1.3309,
      "step": 4382
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.365864494631618e-05,
      "loss": 1.3108,
      "step": 4383
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3657164013328399e-05,
      "loss": 1.4179,
      "step": 4384
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3655683080340617e-05,
      "loss": 1.5046,
      "step": 4385
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3654202147352832e-05,
      "loss": 1.3027,
      "step": 4386
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.365272121436505e-05,
      "loss": 1.471,
      "step": 4387
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3651240281377269e-05,
      "loss": 1.2868,
      "step": 4388
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3649759348389487e-05,
      "loss": 1.209,
      "step": 4389
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3648278415401704e-05,
      "loss": 1.3906,
      "step": 4390
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3646797482413923e-05,
      "loss": 1.1681,
      "step": 4391
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3645316549426139e-05,
      "loss": 1.3337,
      "step": 4392
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3643835616438357e-05,
      "loss": 1.3791,
      "step": 4393
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3642354683450576e-05,
      "loss": 1.347,
      "step": 4394
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3640873750462793e-05,
      "loss": 1.3614,
      "step": 4395
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3639392817475011e-05,
      "loss": 1.4747,
      "step": 4396
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3637911884487227e-05,
      "loss": 1.4898,
      "step": 4397
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3636430951499446e-05,
      "loss": 1.4868,
      "step": 4398
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3634950018511663e-05,
      "loss": 1.6236,
      "step": 4399
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3633469085523881e-05,
      "loss": 1.597,
      "step": 4400
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.36319881525361e-05,
      "loss": 1.6146,
      "step": 4401
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3630507219548316e-05,
      "loss": 1.566,
      "step": 4402
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3629026286560534e-05,
      "loss": 1.3881,
      "step": 4403
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3627545353572751e-05,
      "loss": 1.2722,
      "step": 4404
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.362606442058497e-05,
      "loss": 1.4556,
      "step": 4405
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3624583487597188e-05,
      "loss": 1.2492,
      "step": 4406
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3623102554609404e-05,
      "loss": 1.4943,
      "step": 4407
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3621621621621623e-05,
      "loss": 1.3011,
      "step": 4408
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.362014068863384e-05,
      "loss": 1.5576,
      "step": 4409
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3618659755646058e-05,
      "loss": 1.4673,
      "step": 4410
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3617178822658277e-05,
      "loss": 1.4891,
      "step": 4411
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3615697889670495e-05,
      "loss": 1.47,
      "step": 4412
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.361421695668271e-05,
      "loss": 1.4183,
      "step": 4413
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3612736023694928e-05,
      "loss": 1.3264,
      "step": 4414
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3611255090707147e-05,
      "loss": 1.3402,
      "step": 4415
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3609774157719365e-05,
      "loss": 1.398,
      "step": 4416
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3608293224731582e-05,
      "loss": 1.8335,
      "step": 4417
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3606812291743801e-05,
      "loss": 1.2544,
      "step": 4418
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3605331358756017e-05,
      "loss": 1.5057,
      "step": 4419
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3603850425768235e-05,
      "loss": 1.3771,
      "step": 4420
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.3602369492780452e-05,
      "loss": 1.4271,
      "step": 4421
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3600888559792671e-05,
      "loss": 1.5273,
      "step": 4422
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3599407626804887e-05,
      "loss": 1.383,
      "step": 4423
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3597926693817105e-05,
      "loss": 1.2891,
      "step": 4424
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3596445760829324e-05,
      "loss": 1.3764,
      "step": 4425
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3594964827841541e-05,
      "loss": 1.6492,
      "step": 4426
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3593483894853759e-05,
      "loss": 1.6343,
      "step": 4427
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3592002961865978e-05,
      "loss": 1.4839,
      "step": 4428
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3590522028878194e-05,
      "loss": 1.4086,
      "step": 4429
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3589041095890412e-05,
      "loss": 1.2401,
      "step": 4430
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3587560162902629e-05,
      "loss": 1.4685,
      "step": 4431
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3586079229914848e-05,
      "loss": 1.4056,
      "step": 4432
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3584598296927066e-05,
      "loss": 1.3708,
      "step": 4433
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3583117363939282e-05,
      "loss": 1.5797,
      "step": 4434
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.35816364309515e-05,
      "loss": 1.5817,
      "step": 4435
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3580155497963718e-05,
      "loss": 1.5494,
      "step": 4436
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3578674564975936e-05,
      "loss": 1.5515,
      "step": 4437
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3577193631988153e-05,
      "loss": 1.5083,
      "step": 4438
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3575712699000373e-05,
      "loss": 1.2728,
      "step": 4439
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3574231766012588e-05,
      "loss": 1.5312,
      "step": 4440
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3572750833024806e-05,
      "loss": 1.3813,
      "step": 4441
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3571269900037025e-05,
      "loss": 1.316,
      "step": 4442
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3569788967049243e-05,
      "loss": 1.3376,
      "step": 4443
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3568308034061458e-05,
      "loss": 1.1238,
      "step": 4444
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3566827101073678e-05,
      "loss": 1.4376,
      "step": 4445
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3565346168085895e-05,
      "loss": 1.6438,
      "step": 4446
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3563865235098113e-05,
      "loss": 1.4823,
      "step": 4447
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.356238430211033e-05,
      "loss": 1.2842,
      "step": 4448
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.356090336912255e-05,
      "loss": 1.4335,
      "step": 4449
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3559422436134765e-05,
      "loss": 1.5897,
      "step": 4450
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3557941503146983e-05,
      "loss": 1.4287,
      "step": 4451
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3556460570159202e-05,
      "loss": 1.4378,
      "step": 4452
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.355497963717142e-05,
      "loss": 1.3216,
      "step": 4453
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3553498704183637e-05,
      "loss": 1.4629,
      "step": 4454
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3552017771195856e-05,
      "loss": 1.544,
      "step": 4455
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3550536838208072e-05,
      "loss": 1.5467,
      "step": 4456
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.354905590522029e-05,
      "loss": 1.269,
      "step": 4457
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3547574972232507e-05,
      "loss": 1.5757,
      "step": 4458
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3546094039244726e-05,
      "loss": 1.411,
      "step": 4459
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3544613106256944e-05,
      "loss": 1.3676,
      "step": 4460
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.354313217326916e-05,
      "loss": 1.2411,
      "step": 4461
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3541651240281379e-05,
      "loss": 1.2801,
      "step": 4462
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3540170307293596e-05,
      "loss": 1.6308,
      "step": 4463
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3538689374305814e-05,
      "loss": 1.59,
      "step": 4464
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.353720844131803e-05,
      "loss": 1.2909,
      "step": 4465
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3535727508330249e-05,
      "loss": 1.3042,
      "step": 4466
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3534246575342466e-05,
      "loss": 1.4991,
      "step": 4467
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3532765642354684e-05,
      "loss": 1.3089,
      "step": 4468
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3531284709366903e-05,
      "loss": 1.4472,
      "step": 4469
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.352980377637912e-05,
      "loss": 1.4056,
      "step": 4470
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3528322843391337e-05,
      "loss": 1.536,
      "step": 4471
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3526841910403556e-05,
      "loss": 1.2912,
      "step": 4472
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3525360977415773e-05,
      "loss": 1.303,
      "step": 4473
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.352388004442799e-05,
      "loss": 1.4546,
      "step": 4474
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3522399111440208e-05,
      "loss": 1.5455,
      "step": 4475
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3520918178452427e-05,
      "loss": 1.1936,
      "step": 4476
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3519437245464643e-05,
      "loss": 1.3626,
      "step": 4477
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.351795631247686e-05,
      "loss": 1.3965,
      "step": 4478
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.351647537948908e-05,
      "loss": 1.4001,
      "step": 4479
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3514994446501298e-05,
      "loss": 1.3065,
      "step": 4480
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3513513513513515e-05,
      "loss": 1.3973,
      "step": 4481
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3512032580525731e-05,
      "loss": 1.543,
      "step": 4482
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.351055164753795e-05,
      "loss": 1.3234,
      "step": 4483
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3509070714550168e-05,
      "loss": 1.5565,
      "step": 4484
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3507589781562385e-05,
      "loss": 1.6708,
      "step": 4485
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3506108848574604e-05,
      "loss": 1.2141,
      "step": 4486
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.350462791558682e-05,
      "loss": 1.7268,
      "step": 4487
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3503146982599038e-05,
      "loss": 1.4135,
      "step": 4488
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3501666049611257e-05,
      "loss": 1.2507,
      "step": 4489
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3500185116623474e-05,
      "loss": 1.4895,
      "step": 4490
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3498704183635692e-05,
      "loss": 1.2454,
      "step": 4491
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3497223250647908e-05,
      "loss": 1.5183,
      "step": 4492
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3495742317660127e-05,
      "loss": 1.3687,
      "step": 4493
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3494261384672344e-05,
      "loss": 1.284,
      "step": 4494
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3492780451684562e-05,
      "loss": 1.648,
      "step": 4495
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3491299518696781e-05,
      "loss": 1.4048,
      "step": 4496
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3489818585708999e-05,
      "loss": 1.638,
      "step": 4497
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3488337652721215e-05,
      "loss": 1.468,
      "step": 4498
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3486856719733432e-05,
      "loss": 1.47,
      "step": 4499
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3485375786745651e-05,
      "loss": 1.5739,
      "step": 4500
    },
    {
      "epoch": 0.99,
      "eval_loss": NaN,
      "eval_runtime": 4.3743,
      "eval_samples_per_second": 352.974,
      "eval_steps_per_second": 44.122,
      "step": 4500
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3483894853757869e-05,
      "loss": 1.29,
      "step": 4501
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3482413920770086e-05,
      "loss": 1.441,
      "step": 4502
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3480932987782306e-05,
      "loss": 1.6436,
      "step": 4503
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3479452054794521e-05,
      "loss": 1.4644,
      "step": 4504
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3477971121806739e-05,
      "loss": 1.5006,
      "step": 4505
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3476490188818958e-05,
      "loss": 1.3513,
      "step": 4506
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3475009255831176e-05,
      "loss": 1.1731,
      "step": 4507
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3473528322843391e-05,
      "loss": 1.084,
      "step": 4508
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3472047389855609e-05,
      "loss": 1.5005,
      "step": 4509
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3470566456867828e-05,
      "loss": 1.644,
      "step": 4510
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3469085523880046e-05,
      "loss": 1.6308,
      "step": 4511
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.3467604590892263e-05,
      "loss": 1.5784,
      "step": 4512
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3466123657904482e-05,
      "loss": 1.3733,
      "step": 4513
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3464642724916698e-05,
      "loss": 1.3699,
      "step": 4514
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3463161791928916e-05,
      "loss": 1.5403,
      "step": 4515
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3461680858941133e-05,
      "loss": 1.3295,
      "step": 4516
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3460199925953352e-05,
      "loss": 1.417,
      "step": 4517
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.345871899296557e-05,
      "loss": 1.4203,
      "step": 4518
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3457238059977786e-05,
      "loss": 1.3585,
      "step": 4519
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3455757126990005e-05,
      "loss": 1.295,
      "step": 4520
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3454276194002223e-05,
      "loss": 1.2613,
      "step": 4521
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.345279526101444e-05,
      "loss": 1.3984,
      "step": 4522
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.345131432802666e-05,
      "loss": 1.5185,
      "step": 4523
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3449833395038877e-05,
      "loss": 1.3234,
      "step": 4524
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3448352462051093e-05,
      "loss": 1.3656,
      "step": 4525
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.344687152906331e-05,
      "loss": 1.4605,
      "step": 4526
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.344539059607553e-05,
      "loss": 1.2748,
      "step": 4527
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3443909663087747e-05,
      "loss": 1.629,
      "step": 4528
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3442428730099963e-05,
      "loss": 1.2316,
      "step": 4529
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3440947797112182e-05,
      "loss": 1.4161,
      "step": 4530
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.34394668641244e-05,
      "loss": 1.5333,
      "step": 4531
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3437985931136617e-05,
      "loss": 1.0768,
      "step": 4532
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3436504998148834e-05,
      "loss": 1.9087,
      "step": 4533
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3435024065161054e-05,
      "loss": 1.2858,
      "step": 4534
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.343354313217327e-05,
      "loss": 1.5235,
      "step": 4535
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3432062199185487e-05,
      "loss": 1.3102,
      "step": 4536
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3430581266197706e-05,
      "loss": 1.2505,
      "step": 4537
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3429100333209924e-05,
      "loss": 1.2775,
      "step": 4538
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3427619400222141e-05,
      "loss": 1.4351,
      "step": 4539
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.342613846723436e-05,
      "loss": 1.1666,
      "step": 4540
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3424657534246576e-05,
      "loss": 1.1626,
      "step": 4541
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3423176601258794e-05,
      "loss": 1.5048,
      "step": 4542
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3421695668271011e-05,
      "loss": 1.2855,
      "step": 4543
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.342021473528323e-05,
      "loss": 1.2369,
      "step": 4544
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3418733802295448e-05,
      "loss": 1.3545,
      "step": 4545
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3417252869307664e-05,
      "loss": 1.2804,
      "step": 4546
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3415771936319883e-05,
      "loss": 1.3988,
      "step": 4547
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.34142910033321e-05,
      "loss": 1.5243,
      "step": 4548
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3412810070344318e-05,
      "loss": 1.3371,
      "step": 4549
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3411329137356534e-05,
      "loss": 1.5079,
      "step": 4550
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3409848204368753e-05,
      "loss": 1.3245,
      "step": 4551
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.340836727138097e-05,
      "loss": 1.2565,
      "step": 4552
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3406886338393188e-05,
      "loss": 1.2554,
      "step": 4553
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3405405405405407e-05,
      "loss": 1.1824,
      "step": 4554
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3403924472417625e-05,
      "loss": 1.5672,
      "step": 4555
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.340244353942984e-05,
      "loss": 1.5536,
      "step": 4556
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.340096260644206e-05,
      "loss": 1.4439,
      "step": 4557
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3399481673454277e-05,
      "loss": 1.3553,
      "step": 4558
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3398000740466495e-05,
      "loss": 1.5222,
      "step": 4559
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3396519807478712e-05,
      "loss": 1.3797,
      "step": 4560
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3395038874490932e-05,
      "loss": 1.1432,
      "step": 4561
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3393557941503147e-05,
      "loss": 1.5103,
      "step": 4562
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3392077008515365e-05,
      "loss": 1.0819,
      "step": 4563
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3390596075527584e-05,
      "loss": 1.4663,
      "step": 4564
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3389115142539802e-05,
      "loss": 1.4602,
      "step": 4565
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.338763420955202e-05,
      "loss": 1.0857,
      "step": 4566
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3386153276564235e-05,
      "loss": 1.1037,
      "step": 4567
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3384672343576454e-05,
      "loss": 1.2566,
      "step": 4568
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3383191410588672e-05,
      "loss": 1.2562,
      "step": 4569
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.338171047760089e-05,
      "loss": 1.3835,
      "step": 4570
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3380229544613109e-05,
      "loss": 1.145,
      "step": 4571
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3378748611625324e-05,
      "loss": 1.2859,
      "step": 4572
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3377267678637542e-05,
      "loss": 1.4131,
      "step": 4573
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3375786745649761e-05,
      "loss": 1.3773,
      "step": 4574
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3374305812661979e-05,
      "loss": 1.3793,
      "step": 4575
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3372824879674196e-05,
      "loss": 1.397,
      "step": 4576
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3371343946686412e-05,
      "loss": 1.4428,
      "step": 4577
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3369863013698631e-05,
      "loss": 1.3671,
      "step": 4578
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3368382080710849e-05,
      "loss": 1.2107,
      "step": 4579
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3366901147723066e-05,
      "loss": 1.3677,
      "step": 4580
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3365420214735285e-05,
      "loss": 1.3082,
      "step": 4581
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3363939281747503e-05,
      "loss": 1.1743,
      "step": 4582
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3362458348759719e-05,
      "loss": 1.3342,
      "step": 4583
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3360977415771938e-05,
      "loss": 1.4862,
      "step": 4584
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3359496482784155e-05,
      "loss": 1.4048,
      "step": 4585
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3358015549796373e-05,
      "loss": 1.2076,
      "step": 4586
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3356534616808589e-05,
      "loss": 1.1659,
      "step": 4587
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.335505368382081e-05,
      "loss": 1.3321,
      "step": 4588
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3353572750833026e-05,
      "loss": 1.2926,
      "step": 4589
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3352091817845243e-05,
      "loss": 1.1275,
      "step": 4590
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3350610884857462e-05,
      "loss": 1.1634,
      "step": 4591
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.334912995186968e-05,
      "loss": 1.3574,
      "step": 4592
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3347649018881896e-05,
      "loss": 1.194,
      "step": 4593
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3346168085894113e-05,
      "loss": 1.2272,
      "step": 4594
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3344687152906332e-05,
      "loss": 1.3215,
      "step": 4595
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.334320621991855e-05,
      "loss": 1.1806,
      "step": 4596
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3341725286930767e-05,
      "loss": 1.3698,
      "step": 4597
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3340244353942987e-05,
      "loss": 1.1917,
      "step": 4598
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3338763420955202e-05,
      "loss": 1.4601,
      "step": 4599
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.333728248796742e-05,
      "loss": 1.3288,
      "step": 4600
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3335801554979639e-05,
      "loss": 1.2691,
      "step": 4601
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3334320621991857e-05,
      "loss": 1.4357,
      "step": 4602
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.3332839689004074e-05,
      "loss": 1.2565,
      "step": 4603
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.333135875601629e-05,
      "loss": 1.3863,
      "step": 4604
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.332987782302851e-05,
      "loss": 1.3517,
      "step": 4605
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3328396890040727e-05,
      "loss": 1.33,
      "step": 4606
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3326915957052944e-05,
      "loss": 1.2408,
      "step": 4607
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3325435024065163e-05,
      "loss": 1.4213,
      "step": 4608
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3323954091077381e-05,
      "loss": 1.3092,
      "step": 4609
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3322473158089597e-05,
      "loss": 1.4473,
      "step": 4610
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3320992225101814e-05,
      "loss": 1.2269,
      "step": 4611
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3319511292114034e-05,
      "loss": 1.4241,
      "step": 4612
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3318030359126251e-05,
      "loss": 1.2819,
      "step": 4613
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3316549426138467e-05,
      "loss": 1.4319,
      "step": 4614
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3315068493150686e-05,
      "loss": 1.616,
      "step": 4615
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3313587560162904e-05,
      "loss": 1.4879,
      "step": 4616
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3312106627175121e-05,
      "loss": 1.3293,
      "step": 4617
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.331062569418734e-05,
      "loss": 1.3576,
      "step": 4618
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3309144761199558e-05,
      "loss": 1.4203,
      "step": 4619
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3307663828211774e-05,
      "loss": 1.4364,
      "step": 4620
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3306182895223991e-05,
      "loss": 1.3723,
      "step": 4621
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.330470196223621e-05,
      "loss": 1.3699,
      "step": 4622
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3303221029248428e-05,
      "loss": 1.3346,
      "step": 4623
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3301740096260645e-05,
      "loss": 1.2752,
      "step": 4624
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3300259163272865e-05,
      "loss": 1.3732,
      "step": 4625
    },
    {
      "epoch": 1.02,
      "eval_loss": NaN,
      "eval_runtime": 4.3705,
      "eval_samples_per_second": 353.275,
      "eval_steps_per_second": 44.159,
      "step": 4625
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.329877823028508e-05,
      "loss": 1.3212,
      "step": 4626
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3297297297297298e-05,
      "loss": 1.3853,
      "step": 4627
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3295816364309515e-05,
      "loss": 1.2494,
      "step": 4628
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3294335431321735e-05,
      "loss": 1.1342,
      "step": 4629
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3292854498333952e-05,
      "loss": 1.6207,
      "step": 4630
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3291373565346168e-05,
      "loss": 1.2589,
      "step": 4631
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3289892632358387e-05,
      "loss": 1.5932,
      "step": 4632
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3288411699370605e-05,
      "loss": 1.2047,
      "step": 4633
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3286930766382822e-05,
      "loss": 1.429,
      "step": 4634
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3285449833395041e-05,
      "loss": 1.3321,
      "step": 4635
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3283968900407257e-05,
      "loss": 1.2439,
      "step": 4636
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3282487967419475e-05,
      "loss": 1.2539,
      "step": 4637
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3281007034431692e-05,
      "loss": 1.2399,
      "step": 4638
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3279526101443912e-05,
      "loss": 1.3717,
      "step": 4639
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3278045168456129e-05,
      "loss": 1.4455,
      "step": 4640
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3276564235468345e-05,
      "loss": 1.3981,
      "step": 4641
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3275083302480564e-05,
      "loss": 1.2387,
      "step": 4642
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3273602369492782e-05,
      "loss": 1.1393,
      "step": 4643
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3272121436504999e-05,
      "loss": 1.5663,
      "step": 4644
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3270640503517217e-05,
      "loss": 1.2088,
      "step": 4645
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3269159570529436e-05,
      "loss": 1.2351,
      "step": 4646
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.3267678637541652e-05,
      "loss": 1.3609,
      "step": 4647
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.326619770455387e-05,
      "loss": 1.6073,
      "step": 4648
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3264716771566088e-05,
      "loss": 1.3372,
      "step": 4649
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3263235838578306e-05,
      "loss": 1.3443,
      "step": 4650
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3261754905590522e-05,
      "loss": 1.2887,
      "step": 4651
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3260273972602743e-05,
      "loss": 1.1423,
      "step": 4652
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3258793039614958e-05,
      "loss": 1.3438,
      "step": 4653
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3257312106627176e-05,
      "loss": 1.1735,
      "step": 4654
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3255831173639394e-05,
      "loss": 1.2945,
      "step": 4655
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3254350240651613e-05,
      "loss": 1.3858,
      "step": 4656
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3252869307663829e-05,
      "loss": 1.5955,
      "step": 4657
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3251388374676046e-05,
      "loss": 1.3262,
      "step": 4658
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3249907441688265e-05,
      "loss": 1.4627,
      "step": 4659
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3248426508700483e-05,
      "loss": 1.3025,
      "step": 4660
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.32469455757127e-05,
      "loss": 1.149,
      "step": 4661
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3245464642724916e-05,
      "loss": 1.5009,
      "step": 4662
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3243983709737135e-05,
      "loss": 1.2243,
      "step": 4663
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3242502776749353e-05,
      "loss": 1.4301,
      "step": 4664
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.324102184376157e-05,
      "loss": 1.3793,
      "step": 4665
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.323954091077379e-05,
      "loss": 1.2784,
      "step": 4666
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3238059977786007e-05,
      "loss": 1.3617,
      "step": 4667
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3236579044798223e-05,
      "loss": 1.096,
      "step": 4668
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3235098111810442e-05,
      "loss": 1.4747,
      "step": 4669
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.323361717882266e-05,
      "loss": 1.2597,
      "step": 4670
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3232136245834877e-05,
      "loss": 1.2152,
      "step": 4671
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3230655312847093e-05,
      "loss": 1.4538,
      "step": 4672
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3229174379859314e-05,
      "loss": 1.2025,
      "step": 4673
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.322769344687153e-05,
      "loss": 1.2877,
      "step": 4674
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3226212513883747e-05,
      "loss": 1.1146,
      "step": 4675
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3224731580895966e-05,
      "loss": 1.2056,
      "step": 4676
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3223250647908184e-05,
      "loss": 1.3561,
      "step": 4677
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.32217697149204e-05,
      "loss": 1.3442,
      "step": 4678
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3220288781932617e-05,
      "loss": 1.1795,
      "step": 4679
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3218807848944837e-05,
      "loss": 1.2829,
      "step": 4680
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3217326915957054e-05,
      "loss": 1.2315,
      "step": 4681
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3215845982969272e-05,
      "loss": 1.0706,
      "step": 4682
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.321436504998149e-05,
      "loss": 1.3217,
      "step": 4683
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3212884116993707e-05,
      "loss": 1.3112,
      "step": 4684
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3211403184005924e-05,
      "loss": 1.3881,
      "step": 4685
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3209922251018143e-05,
      "loss": 1.341,
      "step": 4686
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.320844131803036e-05,
      "loss": 1.3064,
      "step": 4687
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3206960385042578e-05,
      "loss": 1.2306,
      "step": 4688
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3205479452054794e-05,
      "loss": 1.2653,
      "step": 4689
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3203998519067013e-05,
      "loss": 1.2344,
      "step": 4690
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3202517586079231e-05,
      "loss": 1.5176,
      "step": 4691
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3201036653091448e-05,
      "loss": 1.4486,
      "step": 4692
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.3199555720103668e-05,
      "loss": 1.2397,
      "step": 4693
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3198074787115885e-05,
      "loss": 1.4139,
      "step": 4694
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3196593854128101e-05,
      "loss": 1.3055,
      "step": 4695
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3195112921140318e-05,
      "loss": 1.1094,
      "step": 4696
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3193631988152538e-05,
      "loss": 1.4008,
      "step": 4697
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3192151055164755e-05,
      "loss": 1.2313,
      "step": 4698
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3190670122176971e-05,
      "loss": 1.3364,
      "step": 4699
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.318918918918919e-05,
      "loss": 1.3185,
      "step": 4700
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3187708256201408e-05,
      "loss": 1.2855,
      "step": 4701
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3186227323213625e-05,
      "loss": 1.2542,
      "step": 4702
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3184746390225844e-05,
      "loss": 1.4209,
      "step": 4703
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3183265457238062e-05,
      "loss": 1.636,
      "step": 4704
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3181784524250278e-05,
      "loss": 1.2938,
      "step": 4705
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3180303591262495e-05,
      "loss": 1.5203,
      "step": 4706
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3178822658274715e-05,
      "loss": 1.1706,
      "step": 4707
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3177341725286932e-05,
      "loss": 1.5295,
      "step": 4708
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.317586079229915e-05,
      "loss": 1.3581,
      "step": 4709
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3174379859311369e-05,
      "loss": 1.3097,
      "step": 4710
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3172898926323585e-05,
      "loss": 1.3486,
      "step": 4711
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3171417993335802e-05,
      "loss": 1.4844,
      "step": 4712
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.316993706034802e-05,
      "loss": 1.2864,
      "step": 4713
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3168456127360239e-05,
      "loss": 1.319,
      "step": 4714
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3166975194372455e-05,
      "loss": 1.2003,
      "step": 4715
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3165494261384672e-05,
      "loss": 1.221,
      "step": 4716
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3164013328396891e-05,
      "loss": 1.341,
      "step": 4717
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3162532395409109e-05,
      "loss": 1.3647,
      "step": 4718
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3161051462421326e-05,
      "loss": 1.2238,
      "step": 4719
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3159570529433546e-05,
      "loss": 1.3078,
      "step": 4720
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3158089596445761e-05,
      "loss": 1.5566,
      "step": 4721
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3156608663457979e-05,
      "loss": 1.3736,
      "step": 4722
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3155127730470197e-05,
      "loss": 1.5398,
      "step": 4723
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3153646797482416e-05,
      "loss": 1.1694,
      "step": 4724
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3152165864494633e-05,
      "loss": 1.3491,
      "step": 4725
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3150684931506849e-05,
      "loss": 1.3674,
      "step": 4726
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3149203998519068e-05,
      "loss": 1.3086,
      "step": 4727
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3147723065531286e-05,
      "loss": 1.0853,
      "step": 4728
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3146242132543503e-05,
      "loss": 1.4286,
      "step": 4729
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3144761199555723e-05,
      "loss": 1.528,
      "step": 4730
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.314328026656794e-05,
      "loss": 1.271,
      "step": 4731
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3141799333580156e-05,
      "loss": 1.3327,
      "step": 4732
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3140318400592373e-05,
      "loss": 1.2446,
      "step": 4733
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3138837467604593e-05,
      "loss": 1.3202,
      "step": 4734
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.313735653461681e-05,
      "loss": 1.4842,
      "step": 4735
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3135875601629026e-05,
      "loss": 1.3376,
      "step": 4736
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3134394668641247e-05,
      "loss": 1.1311,
      "step": 4737
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.3132913735653463e-05,
      "loss": 1.4225,
      "step": 4738
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.313143280266568e-05,
      "loss": 1.553,
      "step": 4739
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3129951869677898e-05,
      "loss": 1.2212,
      "step": 4740
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3128470936690117e-05,
      "loss": 1.3299,
      "step": 4741
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3126990003702333e-05,
      "loss": 1.2069,
      "step": 4742
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.312550907071455e-05,
      "loss": 1.257,
      "step": 4743
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.312402813772677e-05,
      "loss": 1.6175,
      "step": 4744
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3122547204738987e-05,
      "loss": 1.1577,
      "step": 4745
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3121066271751204e-05,
      "loss": 1.3653,
      "step": 4746
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3119585338763424e-05,
      "loss": 1.3695,
      "step": 4747
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.311810440577564e-05,
      "loss": 1.1802,
      "step": 4748
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3116623472787857e-05,
      "loss": 1.2019,
      "step": 4749
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3115142539800075e-05,
      "loss": 1.3156,
      "step": 4750
    },
    {
      "epoch": 1.05,
      "eval_loss": NaN,
      "eval_runtime": 4.3747,
      "eval_samples_per_second": 352.942,
      "eval_steps_per_second": 44.118,
      "step": 4750
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3113661606812294e-05,
      "loss": 1.4483,
      "step": 4751
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3112180673824511e-05,
      "loss": 1.5874,
      "step": 4752
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3110699740836727e-05,
      "loss": 1.3284,
      "step": 4753
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3109218807848946e-05,
      "loss": 1.5707,
      "step": 4754
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3107737874861164e-05,
      "loss": 1.4463,
      "step": 4755
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3106256941873381e-05,
      "loss": 1.2706,
      "step": 4756
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3104776008885597e-05,
      "loss": 1.3833,
      "step": 4757
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3103295075897818e-05,
      "loss": 1.3578,
      "step": 4758
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3101814142910034e-05,
      "loss": 1.4541,
      "step": 4759
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3100333209922251e-05,
      "loss": 1.4322,
      "step": 4760
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.309885227693447e-05,
      "loss": 1.3045,
      "step": 4761
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3097371343946688e-05,
      "loss": 1.372,
      "step": 4762
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3095890410958904e-05,
      "loss": 1.2066,
      "step": 4763
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3094409477971123e-05,
      "loss": 1.2418,
      "step": 4764
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.309292854498334e-05,
      "loss": 1.43,
      "step": 4765
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3091447611995558e-05,
      "loss": 1.3763,
      "step": 4766
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3089966679007776e-05,
      "loss": 1.3386,
      "step": 4767
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3088485746019995e-05,
      "loss": 1.3399,
      "step": 4768
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.308700481303221e-05,
      "loss": 1.2391,
      "step": 4769
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3085523880044428e-05,
      "loss": 1.4246,
      "step": 4770
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3084042947056647e-05,
      "loss": 1.2291,
      "step": 4771
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3082562014068865e-05,
      "loss": 1.1652,
      "step": 4772
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3081081081081083e-05,
      "loss": 1.2535,
      "step": 4773
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3079600148093298e-05,
      "loss": 1.2793,
      "step": 4774
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3078119215105518e-05,
      "loss": 1.4161,
      "step": 4775
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3076638282117735e-05,
      "loss": 1.0663,
      "step": 4776
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3075157349129953e-05,
      "loss": 1.4251,
      "step": 4777
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3073676416142172e-05,
      "loss": 1.2197,
      "step": 4778
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3072195483154388e-05,
      "loss": 1.3159,
      "step": 4779
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3070714550166605e-05,
      "loss": 1.5233,
      "step": 4780
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3069233617178824e-05,
      "loss": 1.2729,
      "step": 4781
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3067752684191042e-05,
      "loss": 1.2688,
      "step": 4782
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.306627175120326e-05,
      "loss": 1.302,
      "step": 4783
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.3064790818215475e-05,
      "loss": 1.5808,
      "step": 4784
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3063309885227694e-05,
      "loss": 1.3468,
      "step": 4785
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3061828952239912e-05,
      "loss": 1.4285,
      "step": 4786
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.306034801925213e-05,
      "loss": 1.4881,
      "step": 4787
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3058867086264349e-05,
      "loss": 1.3995,
      "step": 4788
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3057386153276566e-05,
      "loss": 1.2483,
      "step": 4789
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3055905220288782e-05,
      "loss": 1.2629,
      "step": 4790
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3054424287301e-05,
      "loss": 1.2917,
      "step": 4791
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3052943354313219e-05,
      "loss": 1.4206,
      "step": 4792
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3051462421325436e-05,
      "loss": 1.0747,
      "step": 4793
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3049981488337654e-05,
      "loss": 1.3592,
      "step": 4794
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3048500555349873e-05,
      "loss": 1.3445,
      "step": 4795
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3047019622362089e-05,
      "loss": 1.4098,
      "step": 4796
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3045538689374306e-05,
      "loss": 1.3723,
      "step": 4797
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3044057756386526e-05,
      "loss": 1.1869,
      "step": 4798
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3042576823398743e-05,
      "loss": 1.2742,
      "step": 4799
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3041095890410959e-05,
      "loss": 1.4662,
      "step": 4800
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3039614957423176e-05,
      "loss": 1.4778,
      "step": 4801
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3038134024435396e-05,
      "loss": 1.391,
      "step": 4802
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3036653091447613e-05,
      "loss": 1.3888,
      "step": 4803
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.303517215845983e-05,
      "loss": 1.3962,
      "step": 4804
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.303369122547205e-05,
      "loss": 1.4854,
      "step": 4805
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3032210292484266e-05,
      "loss": 1.4933,
      "step": 4806
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3030729359496483e-05,
      "loss": 1.3126,
      "step": 4807
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.30292484265087e-05,
      "loss": 1.3563,
      "step": 4808
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.302776749352092e-05,
      "loss": 1.7244,
      "step": 4809
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3026286560533137e-05,
      "loss": 1.5125,
      "step": 4810
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3024805627545353e-05,
      "loss": 1.2316,
      "step": 4811
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3023324694557572e-05,
      "loss": 1.3992,
      "step": 4812
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.302184376156979e-05,
      "loss": 1.2693,
      "step": 4813
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3020362828582007e-05,
      "loss": 1.3658,
      "step": 4814
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3018881895594227e-05,
      "loss": 1.3961,
      "step": 4815
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3017400962606444e-05,
      "loss": 1.4353,
      "step": 4816
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.301592002961866e-05,
      "loss": 1.3399,
      "step": 4817
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3014439096630878e-05,
      "loss": 1.4389,
      "step": 4818
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3012958163643097e-05,
      "loss": 1.5209,
      "step": 4819
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3011477230655314e-05,
      "loss": 1.4369,
      "step": 4820
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.300999629766753e-05,
      "loss": 1.3868,
      "step": 4821
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3008515364679751e-05,
      "loss": 1.5006,
      "step": 4822
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3007034431691967e-05,
      "loss": 1.162,
      "step": 4823
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3005553498704184e-05,
      "loss": 1.1496,
      "step": 4824
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3004072565716402e-05,
      "loss": 1.4185,
      "step": 4825
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3002591632728621e-05,
      "loss": 1.3754,
      "step": 4826
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.3001110699740837e-05,
      "loss": 1.4384,
      "step": 4827
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.2999629766753054e-05,
      "loss": 1.2791,
      "step": 4828
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.2998148833765274e-05,
      "loss": 1.171,
      "step": 4829
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2996667900777491e-05,
      "loss": 1.3658,
      "step": 4830
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2995186967789709e-05,
      "loss": 1.2671,
      "step": 4831
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2993706034801928e-05,
      "loss": 1.1757,
      "step": 4832
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2992225101814144e-05,
      "loss": 1.3778,
      "step": 4833
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2990744168826361e-05,
      "loss": 1.2055,
      "step": 4834
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2989263235838579e-05,
      "loss": 1.4213,
      "step": 4835
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2987782302850798e-05,
      "loss": 1.2659,
      "step": 4836
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2986301369863015e-05,
      "loss": 1.2965,
      "step": 4837
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2984820436875231e-05,
      "loss": 1.3822,
      "step": 4838
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.298333950388745e-05,
      "loss": 1.1923,
      "step": 4839
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2981858570899668e-05,
      "loss": 1.36,
      "step": 4840
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2980377637911886e-05,
      "loss": 1.4114,
      "step": 4841
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2978896704924101e-05,
      "loss": 1.1347,
      "step": 4842
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.297741577193632e-05,
      "loss": 1.3012,
      "step": 4843
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2975934838948538e-05,
      "loss": 1.1974,
      "step": 4844
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2974453905960756e-05,
      "loss": 1.4094,
      "step": 4845
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2972972972972975e-05,
      "loss": 1.4269,
      "step": 4846
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2971492039985192e-05,
      "loss": 1.2753,
      "step": 4847
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2970011106997408e-05,
      "loss": 1.3825,
      "step": 4848
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2968530174009627e-05,
      "loss": 1.4373,
      "step": 4849
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2967049241021845e-05,
      "loss": 1.3569,
      "step": 4850
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2965568308034062e-05,
      "loss": 1.1475,
      "step": 4851
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.296408737504628e-05,
      "loss": 1.1987,
      "step": 4852
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2962606442058499e-05,
      "loss": 1.15,
      "step": 4853
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2961125509070715e-05,
      "loss": 1.6508,
      "step": 4854
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2959644576082932e-05,
      "loss": 1.3432,
      "step": 4855
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2958163643095152e-05,
      "loss": 1.3022,
      "step": 4856
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.295668271010737e-05,
      "loss": 1.1815,
      "step": 4857
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2955201777119587e-05,
      "loss": 1.3381,
      "step": 4858
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2953720844131806e-05,
      "loss": 1.1371,
      "step": 4859
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2952239911144022e-05,
      "loss": 1.3353,
      "step": 4860
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.295075897815624e-05,
      "loss": 1.3941,
      "step": 4861
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2949278045168457e-05,
      "loss": 1.2817,
      "step": 4862
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2947797112180676e-05,
      "loss": 1.2195,
      "step": 4863
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2946316179192892e-05,
      "loss": 1.5482,
      "step": 4864
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.294483524620511e-05,
      "loss": 1.4085,
      "step": 4865
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2943354313217329e-05,
      "loss": 1.3046,
      "step": 4866
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2941873380229546e-05,
      "loss": 1.3645,
      "step": 4867
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2940392447241764e-05,
      "loss": 1.3452,
      "step": 4868
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.293891151425398e-05,
      "loss": 1.4654,
      "step": 4869
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2937430581266199e-05,
      "loss": 1.1293,
      "step": 4870
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2935949648278416e-05,
      "loss": 1.3102,
      "step": 4871
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2934468715290634e-05,
      "loss": 1.1855,
      "step": 4872
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2932987782302853e-05,
      "loss": 1.4179,
      "step": 4873
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.293150684931507e-05,
      "loss": 1.2611,
      "step": 4874
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.2930025916327286e-05,
      "loss": 1.1814,
      "step": 4875
    },
    {
      "epoch": 1.07,
      "eval_loss": NaN,
      "eval_runtime": 4.3659,
      "eval_samples_per_second": 353.647,
      "eval_steps_per_second": 44.206,
      "step": 4875
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2928544983339505e-05,
      "loss": 1.5675,
      "step": 4876
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2927064050351723e-05,
      "loss": 1.4278,
      "step": 4877
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.292558311736394e-05,
      "loss": 1.7835,
      "step": 4878
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2924102184376158e-05,
      "loss": 1.2366,
      "step": 4879
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2922621251388377e-05,
      "loss": 1.1533,
      "step": 4880
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2921140318400593e-05,
      "loss": 1.1609,
      "step": 4881
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.291965938541281e-05,
      "loss": 1.3543,
      "step": 4882
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.291817845242503e-05,
      "loss": 1.312,
      "step": 4883
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2916697519437247e-05,
      "loss": 1.2392,
      "step": 4884
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2915216586449463e-05,
      "loss": 1.3143,
      "step": 4885
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.291373565346168e-05,
      "loss": 1.3817,
      "step": 4886
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.29122547204739e-05,
      "loss": 1.3118,
      "step": 4887
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2910773787486117e-05,
      "loss": 1.2649,
      "step": 4888
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2909292854498335e-05,
      "loss": 1.1662,
      "step": 4889
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2907811921510554e-05,
      "loss": 1.4551,
      "step": 4890
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.290633098852277e-05,
      "loss": 1.1135,
      "step": 4891
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2904850055534987e-05,
      "loss": 1.1957,
      "step": 4892
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2903369122547207e-05,
      "loss": 1.5188,
      "step": 4893
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2901888189559424e-05,
      "loss": 1.3082,
      "step": 4894
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2900407256571642e-05,
      "loss": 1.3084,
      "step": 4895
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2898926323583857e-05,
      "loss": 1.2683,
      "step": 4896
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2897445390596077e-05,
      "loss": 1.2023,
      "step": 4897
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2895964457608294e-05,
      "loss": 1.3111,
      "step": 4898
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2894483524620512e-05,
      "loss": 1.1776,
      "step": 4899
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2893002591632731e-05,
      "loss": 1.4256,
      "step": 4900
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2891521658644948e-05,
      "loss": 1.4194,
      "step": 4901
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2890040725657164e-05,
      "loss": 1.1499,
      "step": 4902
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2888559792669382e-05,
      "loss": 1.4018,
      "step": 4903
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2887078859681601e-05,
      "loss": 1.2923,
      "step": 4904
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2885597926693818e-05,
      "loss": 1.2826,
      "step": 4905
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2884116993706034e-05,
      "loss": 1.3532,
      "step": 4906
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2882636060718254e-05,
      "loss": 1.3047,
      "step": 4907
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2881155127730471e-05,
      "loss": 1.3241,
      "step": 4908
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2879674194742689e-05,
      "loss": 1.2101,
      "step": 4909
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2878193261754908e-05,
      "loss": 1.4996,
      "step": 4910
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2876712328767125e-05,
      "loss": 1.3182,
      "step": 4911
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2875231395779341e-05,
      "loss": 1.2845,
      "step": 4912
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2873750462791559e-05,
      "loss": 1.4392,
      "step": 4913
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2872269529803778e-05,
      "loss": 1.6019,
      "step": 4914
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2870788596815995e-05,
      "loss": 1.4504,
      "step": 4915
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2869307663828213e-05,
      "loss": 1.1843,
      "step": 4916
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2867826730840432e-05,
      "loss": 1.142,
      "step": 4917
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2866345797852648e-05,
      "loss": 1.1997,
      "step": 4918
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2864864864864865e-05,
      "loss": 1.2015,
      "step": 4919
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.2863383931877083e-05,
      "loss": 1.2271,
      "step": 4920
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2861902998889302e-05,
      "loss": 1.3509,
      "step": 4921
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.286042206590152e-05,
      "loss": 1.1165,
      "step": 4922
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2858941132913735e-05,
      "loss": 1.1856,
      "step": 4923
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2857460199925955e-05,
      "loss": 1.2574,
      "step": 4924
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2855979266938172e-05,
      "loss": 1.2576,
      "step": 4925
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.285449833395039e-05,
      "loss": 1.4212,
      "step": 4926
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2853017400962609e-05,
      "loss": 1.4677,
      "step": 4927
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2851536467974825e-05,
      "loss": 1.3208,
      "step": 4928
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2850055534987042e-05,
      "loss": 1.2524,
      "step": 4929
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.284857460199926e-05,
      "loss": 1.0999,
      "step": 4930
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2847093669011479e-05,
      "loss": 1.4681,
      "step": 4931
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2845612736023697e-05,
      "loss": 1.2091,
      "step": 4932
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2844131803035912e-05,
      "loss": 1.3646,
      "step": 4933
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2842650870048132e-05,
      "loss": 1.241,
      "step": 4934
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2841169937060349e-05,
      "loss": 1.3247,
      "step": 4935
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2839689004072567e-05,
      "loss": 1.3918,
      "step": 4936
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2838208071084784e-05,
      "loss": 1.3412,
      "step": 4937
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2836727138097003e-05,
      "loss": 1.3575,
      "step": 4938
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2835246205109219e-05,
      "loss": 1.3503,
      "step": 4939
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2833765272121437e-05,
      "loss": 1.2729,
      "step": 4940
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2832284339133656e-05,
      "loss": 1.2694,
      "step": 4941
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2830803406145873e-05,
      "loss": 1.3729,
      "step": 4942
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2829322473158091e-05,
      "loss": 1.1943,
      "step": 4943
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.282784154017031e-05,
      "loss": 1.3178,
      "step": 4944
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2826360607182526e-05,
      "loss": 1.3924,
      "step": 4945
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2824879674194743e-05,
      "loss": 1.2253,
      "step": 4946
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2823398741206961e-05,
      "loss": 1.9203,
      "step": 4947
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.282191780821918e-05,
      "loss": 1.2747,
      "step": 4948
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2820436875231396e-05,
      "loss": 1.4816,
      "step": 4949
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2818955942243614e-05,
      "loss": 1.3973,
      "step": 4950
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2817475009255833e-05,
      "loss": 1.343,
      "step": 4951
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.281599407626805e-05,
      "loss": 1.5653,
      "step": 4952
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2814513143280268e-05,
      "loss": 1.6368,
      "step": 4953
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2813032210292484e-05,
      "loss": 1.2196,
      "step": 4954
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2811551277304703e-05,
      "loss": 1.2884,
      "step": 4955
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.281007034431692e-05,
      "loss": 1.2742,
      "step": 4956
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2808589411329138e-05,
      "loss": 1.5312,
      "step": 4957
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2807108478341357e-05,
      "loss": 1.4367,
      "step": 4958
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2805627545353575e-05,
      "loss": 1.2877,
      "step": 4959
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.280414661236579e-05,
      "loss": 1.0787,
      "step": 4960
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.280266567937801e-05,
      "loss": 1.3257,
      "step": 4961
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2801184746390227e-05,
      "loss": 1.2041,
      "step": 4962
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2799703813402445e-05,
      "loss": 1.2866,
      "step": 4963
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2798222880414662e-05,
      "loss": 1.4586,
      "step": 4964
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.2796741947426881e-05,
      "loss": 1.3553,
      "step": 4965
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2795261014439097e-05,
      "loss": 1.3979,
      "step": 4966
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2793780081451315e-05,
      "loss": 1.1576,
      "step": 4967
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2792299148463534e-05,
      "loss": 1.3563,
      "step": 4968
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2790818215475751e-05,
      "loss": 1.4699,
      "step": 4969
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2789337282487967e-05,
      "loss": 1.5239,
      "step": 4970
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2787856349500185e-05,
      "loss": 1.275,
      "step": 4971
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2786375416512404e-05,
      "loss": 1.2508,
      "step": 4972
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2784894483524621e-05,
      "loss": 1.1338,
      "step": 4973
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2783413550536839e-05,
      "loss": 1.4464,
      "step": 4974
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2781932617549058e-05,
      "loss": 1.5974,
      "step": 4975
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2780451684561274e-05,
      "loss": 1.5495,
      "step": 4976
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2778970751573492e-05,
      "loss": 1.4321,
      "step": 4977
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.277748981858571e-05,
      "loss": 1.4788,
      "step": 4978
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2776008885597928e-05,
      "loss": 1.4879,
      "step": 4979
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2774527952610146e-05,
      "loss": 1.4057,
      "step": 4980
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2773047019622362e-05,
      "loss": 1.1118,
      "step": 4981
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.277156608663458e-05,
      "loss": 1.1262,
      "step": 4982
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2770085153646798e-05,
      "loss": 1.2598,
      "step": 4983
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2768604220659016e-05,
      "loss": 1.3977,
      "step": 4984
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2767123287671235e-05,
      "loss": 1.2791,
      "step": 4985
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2765642354683453e-05,
      "loss": 1.3481,
      "step": 4986
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2764161421695668e-05,
      "loss": 1.4538,
      "step": 4987
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2762680488707888e-05,
      "loss": 1.1413,
      "step": 4988
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2761199555720105e-05,
      "loss": 1.3385,
      "step": 4989
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2759718622732323e-05,
      "loss": 1.3048,
      "step": 4990
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2758237689744538e-05,
      "loss": 1.4235,
      "step": 4991
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2756756756756758e-05,
      "loss": 1.2799,
      "step": 4992
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2755275823768975e-05,
      "loss": 1.3808,
      "step": 4993
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2753794890781193e-05,
      "loss": 1.2344,
      "step": 4994
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2752313957793412e-05,
      "loss": 1.5101,
      "step": 4995
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.275083302480563e-05,
      "loss": 1.0667,
      "step": 4996
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2749352091817845e-05,
      "loss": 1.3233,
      "step": 4997
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2747871158830063e-05,
      "loss": 1.3062,
      "step": 4998
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2746390225842282e-05,
      "loss": 1.592,
      "step": 4999
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.27449092928545e-05,
      "loss": 1.391,
      "step": 5000
    },
    {
      "epoch": 1.1,
      "eval_loss": NaN,
      "eval_runtime": 4.3764,
      "eval_samples_per_second": 352.805,
      "eval_steps_per_second": 44.101,
      "step": 5000
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2743428359866717e-05,
      "loss": 1.2597,
      "step": 5001
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2741947426878936e-05,
      "loss": 1.184,
      "step": 5002
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2740466493891152e-05,
      "loss": 1.3244,
      "step": 5003
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.273898556090337e-05,
      "loss": 1.2832,
      "step": 5004
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2737504627915589e-05,
      "loss": 1.2747,
      "step": 5005
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2736023694927806e-05,
      "loss": 1.1547,
      "step": 5006
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2734542761940024e-05,
      "loss": 1.207,
      "step": 5007
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.273306182895224e-05,
      "loss": 1.4899,
      "step": 5008
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2731580895964459e-05,
      "loss": 1.305,
      "step": 5009
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2730099962976676e-05,
      "loss": 1.4559,
      "step": 5010
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.2728619029988894e-05,
      "loss": 1.2088,
      "step": 5011
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2727138097001113e-05,
      "loss": 1.2095,
      "step": 5012
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2725657164013329e-05,
      "loss": 1.0483,
      "step": 5013
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2724176231025546e-05,
      "loss": 1.2432,
      "step": 5014
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2722695298037764e-05,
      "loss": 1.4494,
      "step": 5015
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2721214365049983e-05,
      "loss": 1.4688,
      "step": 5016
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.27197334320622e-05,
      "loss": 1.3536,
      "step": 5017
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2718252499074417e-05,
      "loss": 1.3925,
      "step": 5018
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2716771566086636e-05,
      "loss": 1.5122,
      "step": 5019
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2715290633098853e-05,
      "loss": 1.2018,
      "step": 5020
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.271380970011107e-05,
      "loss": 1.2156,
      "step": 5021
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.271232876712329e-05,
      "loss": 1.243,
      "step": 5022
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2710847834135507e-05,
      "loss": 1.3404,
      "step": 5023
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2709366901147723e-05,
      "loss": 1.1663,
      "step": 5024
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.270788596815994e-05,
      "loss": 1.3292,
      "step": 5025
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.270640503517216e-05,
      "loss": 1.1683,
      "step": 5026
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2704924102184378e-05,
      "loss": 1.2388,
      "step": 5027
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2703443169196595e-05,
      "loss": 1.363,
      "step": 5028
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2701962236208814e-05,
      "loss": 1.3836,
      "step": 5029
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.270048130322103e-05,
      "loss": 1.3605,
      "step": 5030
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2699000370233248e-05,
      "loss": 1.3172,
      "step": 5031
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2697519437245465e-05,
      "loss": 1.4246,
      "step": 5032
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2696038504257684e-05,
      "loss": 1.3768,
      "step": 5033
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.26945575712699e-05,
      "loss": 1.4684,
      "step": 5034
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2693076638282118e-05,
      "loss": 1.6326,
      "step": 5035
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2691595705294337e-05,
      "loss": 1.412,
      "step": 5036
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2690114772306554e-05,
      "loss": 1.3489,
      "step": 5037
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2688633839318772e-05,
      "loss": 1.2067,
      "step": 5038
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2687152906330991e-05,
      "loss": 1.1489,
      "step": 5039
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2685671973343207e-05,
      "loss": 1.3515,
      "step": 5040
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2684191040355424e-05,
      "loss": 1.5458,
      "step": 5041
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2682710107367642e-05,
      "loss": 1.3495,
      "step": 5042
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2681229174379861e-05,
      "loss": 1.4002,
      "step": 5043
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2679748241392079e-05,
      "loss": 1.6016,
      "step": 5044
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2678267308404295e-05,
      "loss": 1.2042,
      "step": 5045
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2676786375416514e-05,
      "loss": 1.4563,
      "step": 5046
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2675305442428731e-05,
      "loss": 1.2181,
      "step": 5047
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2673824509440949e-05,
      "loss": 1.365,
      "step": 5048
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2672343576453166e-05,
      "loss": 1.4706,
      "step": 5049
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2670862643465386e-05,
      "loss": 1.0728,
      "step": 5050
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2669381710477601e-05,
      "loss": 1.2647,
      "step": 5051
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2667900777489819e-05,
      "loss": 1.3931,
      "step": 5052
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2666419844502038e-05,
      "loss": 1.4023,
      "step": 5053
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2664938911514256e-05,
      "loss": 1.2568,
      "step": 5054
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.2663457978526471e-05,
      "loss": 1.2145,
      "step": 5055
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.266197704553869e-05,
      "loss": 1.1287,
      "step": 5056
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2660496112550908e-05,
      "loss": 1.3212,
      "step": 5057
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2659015179563126e-05,
      "loss": 1.1774,
      "step": 5058
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2657534246575343e-05,
      "loss": 1.2871,
      "step": 5059
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2656053313587562e-05,
      "loss": 1.5476,
      "step": 5060
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2654572380599778e-05,
      "loss": 1.2334,
      "step": 5061
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2653091447611996e-05,
      "loss": 1.2406,
      "step": 5062
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2651610514624215e-05,
      "loss": 1.4268,
      "step": 5063
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2650129581636432e-05,
      "loss": 1.5208,
      "step": 5064
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.264864864864865e-05,
      "loss": 1.3522,
      "step": 5065
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2647167715660866e-05,
      "loss": 1.3654,
      "step": 5066
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2645686782673085e-05,
      "loss": 1.2658,
      "step": 5067
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2644205849685303e-05,
      "loss": 1.2198,
      "step": 5068
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.264272491669752e-05,
      "loss": 1.4396,
      "step": 5069
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.264124398370974e-05,
      "loss": 1.3135,
      "step": 5070
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2639763050721957e-05,
      "loss": 1.3071,
      "step": 5071
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2638282117734173e-05,
      "loss": 1.2075,
      "step": 5072
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2636801184746392e-05,
      "loss": 1.443,
      "step": 5073
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.263532025175861e-05,
      "loss": 1.4144,
      "step": 5074
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2633839318770827e-05,
      "loss": 1.3307,
      "step": 5075
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2632358385783043e-05,
      "loss": 1.2842,
      "step": 5076
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2630877452795262e-05,
      "loss": 1.3684,
      "step": 5077
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.262939651980748e-05,
      "loss": 1.3425,
      "step": 5078
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2627915586819697e-05,
      "loss": 1.391,
      "step": 5079
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2626434653831916e-05,
      "loss": 1.3423,
      "step": 5080
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2624953720844134e-05,
      "loss": 1.3105,
      "step": 5081
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.262347278785635e-05,
      "loss": 1.5069,
      "step": 5082
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2621991854868567e-05,
      "loss": 1.3055,
      "step": 5083
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2620510921880786e-05,
      "loss": 1.0402,
      "step": 5084
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2619029988893004e-05,
      "loss": 1.1944,
      "step": 5085
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2617549055905221e-05,
      "loss": 1.3683,
      "step": 5086
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.261606812291744e-05,
      "loss": 1.3541,
      "step": 5087
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2614587189929656e-05,
      "loss": 1.476,
      "step": 5088
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2613106256941874e-05,
      "loss": 1.288,
      "step": 5089
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2611625323954093e-05,
      "loss": 1.3539,
      "step": 5090
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.261014439096631e-05,
      "loss": 1.4478,
      "step": 5091
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2608663457978528e-05,
      "loss": 1.3063,
      "step": 5092
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2607182524990744e-05,
      "loss": 1.3185,
      "step": 5093
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2605701592002963e-05,
      "loss": 1.2185,
      "step": 5094
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.260422065901518e-05,
      "loss": 1.2856,
      "step": 5095
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2602739726027398e-05,
      "loss": 1.3528,
      "step": 5096
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2601258793039617e-05,
      "loss": 1.4571,
      "step": 5097
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2599777860051833e-05,
      "loss": 1.453,
      "step": 5098
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.259829692706405e-05,
      "loss": 1.2265,
      "step": 5099
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2596815994076268e-05,
      "loss": 1.2741,
      "step": 5100
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.2595335061088487e-05,
      "loss": 1.2808,
      "step": 5101
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2593854128100705e-05,
      "loss": 1.3916,
      "step": 5102
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.259237319511292e-05,
      "loss": 1.3072,
      "step": 5103
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.259089226212514e-05,
      "loss": 1.4164,
      "step": 5104
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2589411329137357e-05,
      "loss": 1.3996,
      "step": 5105
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2587930396149575e-05,
      "loss": 1.3274,
      "step": 5106
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2586449463161794e-05,
      "loss": 1.2652,
      "step": 5107
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2584968530174012e-05,
      "loss": 1.3856,
      "step": 5108
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2583487597186227e-05,
      "loss": 1.3051,
      "step": 5109
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2582006664198445e-05,
      "loss": 1.3918,
      "step": 5110
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2580525731210664e-05,
      "loss": 1.1029,
      "step": 5111
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2579044798222882e-05,
      "loss": 1.1718,
      "step": 5112
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.25775638652351e-05,
      "loss": 1.4869,
      "step": 5113
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2576082932247318e-05,
      "loss": 1.3937,
      "step": 5114
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2574601999259534e-05,
      "loss": 1.2165,
      "step": 5115
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2573121066271752e-05,
      "loss": 1.2503,
      "step": 5116
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2571640133283971e-05,
      "loss": 1.4086,
      "step": 5117
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2570159200296189e-05,
      "loss": 1.2992,
      "step": 5118
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2568678267308404e-05,
      "loss": 1.5448,
      "step": 5119
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2567197334320622e-05,
      "loss": 1.3214,
      "step": 5120
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2565716401332841e-05,
      "loss": 1.3698,
      "step": 5121
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2564235468345059e-05,
      "loss": 1.144,
      "step": 5122
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2562754535357276e-05,
      "loss": 1.2216,
      "step": 5123
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2561273602369495e-05,
      "loss": 1.1857,
      "step": 5124
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2559792669381711e-05,
      "loss": 1.3179,
      "step": 5125
    },
    {
      "epoch": 1.13,
      "eval_loss": NaN,
      "eval_runtime": 4.3751,
      "eval_samples_per_second": 352.907,
      "eval_steps_per_second": 44.113,
      "step": 5125
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2558311736393929e-05,
      "loss": 1.2784,
      "step": 5126
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2556830803406146e-05,
      "loss": 1.4756,
      "step": 5127
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2555349870418365e-05,
      "loss": 1.393,
      "step": 5128
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2553868937430583e-05,
      "loss": 1.5053,
      "step": 5129
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2552388004442799e-05,
      "loss": 1.4623,
      "step": 5130
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2550907071455018e-05,
      "loss": 1.3103,
      "step": 5131
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2549426138467235e-05,
      "loss": 1.3433,
      "step": 5132
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2547945205479453e-05,
      "loss": 1.3396,
      "step": 5133
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2546464272491672e-05,
      "loss": 1.1455,
      "step": 5134
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.254498333950389e-05,
      "loss": 1.3216,
      "step": 5135
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2543502406516106e-05,
      "loss": 1.4917,
      "step": 5136
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2542021473528323e-05,
      "loss": 1.6707,
      "step": 5137
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2540540540540542e-05,
      "loss": 1.1787,
      "step": 5138
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.253905960755276e-05,
      "loss": 1.0341,
      "step": 5139
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2537578674564976e-05,
      "loss": 1.3743,
      "step": 5140
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2536097741577195e-05,
      "loss": 1.5017,
      "step": 5141
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2534616808589412e-05,
      "loss": 1.446,
      "step": 5142
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.253313587560163e-05,
      "loss": 1.4184,
      "step": 5143
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2531654942613847e-05,
      "loss": 1.4204,
      "step": 5144
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2530174009626067e-05,
      "loss": 1.1937,
      "step": 5145
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.2528693076638282e-05,
      "loss": 1.3076,
      "step": 5146
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.25272121436505e-05,
      "loss": 1.2651,
      "step": 5147
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2525731210662719e-05,
      "loss": 1.1927,
      "step": 5148
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2524250277674937e-05,
      "loss": 1.3359,
      "step": 5149
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2522769344687154e-05,
      "loss": 1.3968,
      "step": 5150
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2521288411699373e-05,
      "loss": 1.2172,
      "step": 5151
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.251980747871159e-05,
      "loss": 1.4569,
      "step": 5152
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2518326545723807e-05,
      "loss": 1.1774,
      "step": 5153
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2516845612736024e-05,
      "loss": 1.0725,
      "step": 5154
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2515364679748243e-05,
      "loss": 1.3913,
      "step": 5155
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2513883746760461e-05,
      "loss": 1.3495,
      "step": 5156
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2512402813772677e-05,
      "loss": 1.3986,
      "step": 5157
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2510921880784896e-05,
      "loss": 1.2999,
      "step": 5158
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2509440947797114e-05,
      "loss": 1.2622,
      "step": 5159
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2507960014809331e-05,
      "loss": 1.4007,
      "step": 5160
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2506479081821547e-05,
      "loss": 1.118,
      "step": 5161
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2504998148833766e-05,
      "loss": 1.0974,
      "step": 5162
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2503517215845984e-05,
      "loss": 1.3465,
      "step": 5163
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2502036282858201e-05,
      "loss": 1.2084,
      "step": 5164
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.250055534987042e-05,
      "loss": 1.1532,
      "step": 5165
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2499074416882638e-05,
      "loss": 1.351,
      "step": 5166
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2497593483894854e-05,
      "loss": 1.4515,
      "step": 5167
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2496112550907073e-05,
      "loss": 1.2958,
      "step": 5168
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.249463161791929e-05,
      "loss": 1.4926,
      "step": 5169
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2493150684931508e-05,
      "loss": 1.3392,
      "step": 5170
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2491669751943725e-05,
      "loss": 1.5382,
      "step": 5171
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2490188818955945e-05,
      "loss": 1.2941,
      "step": 5172
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.248870788596816e-05,
      "loss": 1.2144,
      "step": 5173
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2487226952980378e-05,
      "loss": 1.3759,
      "step": 5174
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2485746019992597e-05,
      "loss": 1.182,
      "step": 5175
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2484265087004815e-05,
      "loss": 1.254,
      "step": 5176
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2482784154017032e-05,
      "loss": 1.1277,
      "step": 5177
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2481303221029248e-05,
      "loss": 1.2304,
      "step": 5178
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2479822288041467e-05,
      "loss": 1.4419,
      "step": 5179
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2478341355053685e-05,
      "loss": 1.3439,
      "step": 5180
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2476860422065902e-05,
      "loss": 1.2673,
      "step": 5181
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2475379489078121e-05,
      "loss": 1.3647,
      "step": 5182
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2473898556090337e-05,
      "loss": 1.3463,
      "step": 5183
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2472417623102555e-05,
      "loss": 1.1962,
      "step": 5184
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2470936690114774e-05,
      "loss": 1.2113,
      "step": 5185
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2469455757126992e-05,
      "loss": 1.382,
      "step": 5186
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2467974824139209e-05,
      "loss": 1.6046,
      "step": 5187
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2466493891151425e-05,
      "loss": 1.1741,
      "step": 5188
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2465012958163644e-05,
      "loss": 1.2178,
      "step": 5189
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2463532025175862e-05,
      "loss": 1.4185,
      "step": 5190
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2462051092188079e-05,
      "loss": 1.4475,
      "step": 5191
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.2460570159200298e-05,
      "loss": 1.2218,
      "step": 5192
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2459089226212516e-05,
      "loss": 1.3621,
      "step": 5193
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2457608293224732e-05,
      "loss": 1.4653,
      "step": 5194
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.245612736023695e-05,
      "loss": 1.3026,
      "step": 5195
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2454646427249168e-05,
      "loss": 1.2767,
      "step": 5196
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2453165494261386e-05,
      "loss": 1.7268,
      "step": 5197
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2451684561273602e-05,
      "loss": 1.4828,
      "step": 5198
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2450203628285823e-05,
      "loss": 1.3321,
      "step": 5199
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2448722695298038e-05,
      "loss": 1.3417,
      "step": 5200
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2447241762310256e-05,
      "loss": 1.5043,
      "step": 5201
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2445760829322475e-05,
      "loss": 1.324,
      "step": 5202
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2444279896334693e-05,
      "loss": 1.4679,
      "step": 5203
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2442798963346909e-05,
      "loss": 1.1328,
      "step": 5204
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2441318030359126e-05,
      "loss": 1.1333,
      "step": 5205
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2439837097371345e-05,
      "loss": 1.2402,
      "step": 5206
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2438356164383563e-05,
      "loss": 1.4229,
      "step": 5207
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.243687523139578e-05,
      "loss": 1.2908,
      "step": 5208
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2435394298408e-05,
      "loss": 1.5043,
      "step": 5209
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2433913365420215e-05,
      "loss": 1.3235,
      "step": 5210
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2432432432432433e-05,
      "loss": 1.3036,
      "step": 5211
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.243095149944465e-05,
      "loss": 1.3493,
      "step": 5212
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.242947056645687e-05,
      "loss": 1.1632,
      "step": 5213
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2427989633469087e-05,
      "loss": 1.3481,
      "step": 5214
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2426508700481303e-05,
      "loss": 1.301,
      "step": 5215
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2425027767493522e-05,
      "loss": 1.5672,
      "step": 5216
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.242354683450574e-05,
      "loss": 1.4726,
      "step": 5217
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2422065901517957e-05,
      "loss": 1.5558,
      "step": 5218
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2420584968530176e-05,
      "loss": 1.4098,
      "step": 5219
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2419104035542394e-05,
      "loss": 1.5681,
      "step": 5220
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.241762310255461e-05,
      "loss": 1.3231,
      "step": 5221
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2416142169566827e-05,
      "loss": 1.2371,
      "step": 5222
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2414661236579046e-05,
      "loss": 1.3646,
      "step": 5223
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2413180303591264e-05,
      "loss": 1.2342,
      "step": 5224
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.241169937060348e-05,
      "loss": 1.4469,
      "step": 5225
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2410218437615699e-05,
      "loss": 1.3736,
      "step": 5226
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2408737504627917e-05,
      "loss": 1.2649,
      "step": 5227
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2407256571640134e-05,
      "loss": 1.4259,
      "step": 5228
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2405775638652352e-05,
      "loss": 1.3824,
      "step": 5229
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.240429470566457e-05,
      "loss": 1.2053,
      "step": 5230
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2402813772676787e-05,
      "loss": 1.325,
      "step": 5231
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2401332839689004e-05,
      "loss": 1.3049,
      "step": 5232
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2399851906701223e-05,
      "loss": 1.537,
      "step": 5233
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.239837097371344e-05,
      "loss": 1.2787,
      "step": 5234
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2396890040725658e-05,
      "loss": 1.1255,
      "step": 5235
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2395409107737878e-05,
      "loss": 1.1665,
      "step": 5236
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2393928174750093e-05,
      "loss": 1.2618,
      "step": 5237
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2392447241762311e-05,
      "loss": 1.3677,
      "step": 5238
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2390966308774528e-05,
      "loss": 1.2966,
      "step": 5239
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2389485375786748e-05,
      "loss": 1.294,
      "step": 5240
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2388004442798965e-05,
      "loss": 1.4745,
      "step": 5241
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2386523509811181e-05,
      "loss": 1.6321,
      "step": 5242
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.23850425768234e-05,
      "loss": 1.3947,
      "step": 5243
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2383561643835618e-05,
      "loss": 1.4057,
      "step": 5244
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2382080710847835e-05,
      "loss": 1.0667,
      "step": 5245
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2380599777860051e-05,
      "loss": 1.3831,
      "step": 5246
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.237911884487227e-05,
      "loss": 1.4762,
      "step": 5247
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2377637911884488e-05,
      "loss": 1.0673,
      "step": 5248
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2376156978896705e-05,
      "loss": 1.4594,
      "step": 5249
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2374676045908924e-05,
      "loss": 1.2927,
      "step": 5250
    },
    {
      "epoch": 1.16,
      "eval_loss": NaN,
      "eval_runtime": 4.3742,
      "eval_samples_per_second": 352.975,
      "eval_steps_per_second": 44.122,
      "step": 5250
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2373195112921142e-05,
      "loss": 1.1768,
      "step": 5251
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2371714179933358e-05,
      "loss": 1.5173,
      "step": 5252
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2370233246945577e-05,
      "loss": 1.2835,
      "step": 5253
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2368752313957795e-05,
      "loss": 1.1057,
      "step": 5254
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2367271380970012e-05,
      "loss": 1.348,
      "step": 5255
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.236579044798223e-05,
      "loss": 1.2807,
      "step": 5256
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2364309514994449e-05,
      "loss": 1.3595,
      "step": 5257
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2362828582006665e-05,
      "loss": 1.3188,
      "step": 5258
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2361347649018882e-05,
      "loss": 1.3592,
      "step": 5259
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2359866716031101e-05,
      "loss": 1.265,
      "step": 5260
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2358385783043319e-05,
      "loss": 1.3808,
      "step": 5261
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2356904850055535e-05,
      "loss": 1.3289,
      "step": 5262
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2355423917067756e-05,
      "loss": 1.3428,
      "step": 5263
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2353942984079971e-05,
      "loss": 1.5097,
      "step": 5264
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2352462051092189e-05,
      "loss": 1.3333,
      "step": 5265
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2350981118104406e-05,
      "loss": 1.1614,
      "step": 5266
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2349500185116626e-05,
      "loss": 1.3152,
      "step": 5267
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2348019252128841e-05,
      "loss": 1.082,
      "step": 5268
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2346538319141059e-05,
      "loss": 1.4612,
      "step": 5269
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2345057386153278e-05,
      "loss": 1.2279,
      "step": 5270
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2343576453165496e-05,
      "loss": 1.2282,
      "step": 5271
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2342095520177713e-05,
      "loss": 1.3397,
      "step": 5272
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2340614587189929e-05,
      "loss": 1.6365,
      "step": 5273
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2339133654202148e-05,
      "loss": 1.491,
      "step": 5274
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2337652721214366e-05,
      "loss": 1.2451,
      "step": 5275
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2336171788226583e-05,
      "loss": 1.3146,
      "step": 5276
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2334690855238803e-05,
      "loss": 1.4521,
      "step": 5277
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.233320992225102e-05,
      "loss": 1.5004,
      "step": 5278
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2331728989263236e-05,
      "loss": 1.1118,
      "step": 5279
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2330248056275455e-05,
      "loss": 1.3403,
      "step": 5280
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2328767123287673e-05,
      "loss": 1.5309,
      "step": 5281
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.232728619029989e-05,
      "loss": 1.3878,
      "step": 5282
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.2325805257312106e-05,
      "loss": 1.4244,
      "step": 5283
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2324324324324327e-05,
      "loss": 1.4934,
      "step": 5284
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2322843391336543e-05,
      "loss": 1.3563,
      "step": 5285
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.232136245834876e-05,
      "loss": 1.3255,
      "step": 5286
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.231988152536098e-05,
      "loss": 1.4564,
      "step": 5287
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2318400592373197e-05,
      "loss": 1.3816,
      "step": 5288
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2316919659385413e-05,
      "loss": 1.2826,
      "step": 5289
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.231543872639763e-05,
      "loss": 1.2117,
      "step": 5290
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.231395779340985e-05,
      "loss": 1.2154,
      "step": 5291
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2312476860422067e-05,
      "loss": 1.2718,
      "step": 5292
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2310995927434284e-05,
      "loss": 1.309,
      "step": 5293
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2309514994446504e-05,
      "loss": 1.5255,
      "step": 5294
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.230803406145872e-05,
      "loss": 1.4861,
      "step": 5295
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2306553128470937e-05,
      "loss": 1.5272,
      "step": 5296
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2305072195483156e-05,
      "loss": 1.2238,
      "step": 5297
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2303591262495374e-05,
      "loss": 1.256,
      "step": 5298
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2302110329507591e-05,
      "loss": 1.4972,
      "step": 5299
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2300629396519807e-05,
      "loss": 1.276,
      "step": 5300
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2299148463532026e-05,
      "loss": 1.1582,
      "step": 5301
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2297667530544244e-05,
      "loss": 0.9894,
      "step": 5302
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2296186597556461e-05,
      "loss": 1.4998,
      "step": 5303
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.229470566456868e-05,
      "loss": 1.4255,
      "step": 5304
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2293224731580898e-05,
      "loss": 1.5311,
      "step": 5305
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2291743798593114e-05,
      "loss": 1.174,
      "step": 5306
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2290262865605331e-05,
      "loss": 1.26,
      "step": 5307
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.228878193261755e-05,
      "loss": 1.4018,
      "step": 5308
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2287300999629768e-05,
      "loss": 1.5461,
      "step": 5309
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2285820066641984e-05,
      "loss": 1.3501,
      "step": 5310
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2284339133654203e-05,
      "loss": 1.3468,
      "step": 5311
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.228285820066642e-05,
      "loss": 1.5884,
      "step": 5312
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2281377267678638e-05,
      "loss": 1.2065,
      "step": 5313
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2279896334690857e-05,
      "loss": 1.3066,
      "step": 5314
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2278415401703075e-05,
      "loss": 1.5547,
      "step": 5315
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.227693446871529e-05,
      "loss": 1.643,
      "step": 5316
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2275453535727508e-05,
      "loss": 1.3861,
      "step": 5317
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2273972602739727e-05,
      "loss": 1.2982,
      "step": 5318
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2272491669751945e-05,
      "loss": 1.3506,
      "step": 5319
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2271010736764163e-05,
      "loss": 1.1855,
      "step": 5320
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2269529803776382e-05,
      "loss": 1.3234,
      "step": 5321
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2268048870788598e-05,
      "loss": 1.5028,
      "step": 5322
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2266567937800815e-05,
      "loss": 1.2418,
      "step": 5323
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2265087004813033e-05,
      "loss": 1.3381,
      "step": 5324
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2263606071825252e-05,
      "loss": 1.5417,
      "step": 5325
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2262125138837468e-05,
      "loss": 1.6319,
      "step": 5326
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2260644205849685e-05,
      "loss": 1.3228,
      "step": 5327
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2259163272861904e-05,
      "loss": 1.1699,
      "step": 5328
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2257682339874122e-05,
      "loss": 1.191,
      "step": 5329
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.225620140688634e-05,
      "loss": 1.2861,
      "step": 5330
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2254720473898559e-05,
      "loss": 1.5585,
      "step": 5331
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2253239540910774e-05,
      "loss": 1.3705,
      "step": 5332
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2251758607922992e-05,
      "loss": 1.4446,
      "step": 5333
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.225027767493521e-05,
      "loss": 1.4808,
      "step": 5334
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2248796741947429e-05,
      "loss": 1.4349,
      "step": 5335
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2247315808959646e-05,
      "loss": 1.4068,
      "step": 5336
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2245834875971862e-05,
      "loss": 1.2151,
      "step": 5337
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2244353942984081e-05,
      "loss": 1.2973,
      "step": 5338
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2242873009996299e-05,
      "loss": 1.2906,
      "step": 5339
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2241392077008516e-05,
      "loss": 1.0962,
      "step": 5340
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2239911144020734e-05,
      "loss": 1.4163,
      "step": 5341
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2238430211032953e-05,
      "loss": 1.3903,
      "step": 5342
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2236949278045169e-05,
      "loss": 1.3397,
      "step": 5343
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2235468345057386e-05,
      "loss": 1.2744,
      "step": 5344
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2233987412069606e-05,
      "loss": 1.4206,
      "step": 5345
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2232506479081823e-05,
      "loss": 1.1807,
      "step": 5346
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2231025546094039e-05,
      "loss": 1.2613,
      "step": 5347
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.222954461310626e-05,
      "loss": 1.4341,
      "step": 5348
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2228063680118476e-05,
      "loss": 1.3079,
      "step": 5349
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2226582747130693e-05,
      "loss": 1.4779,
      "step": 5350
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.222510181414291e-05,
      "loss": 1.3562,
      "step": 5351
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.222362088115513e-05,
      "loss": 1.435,
      "step": 5352
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2222139948167346e-05,
      "loss": 1.4697,
      "step": 5353
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2220659015179563e-05,
      "loss": 1.3856,
      "step": 5354
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2219178082191782e-05,
      "loss": 1.2566,
      "step": 5355
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2217697149204e-05,
      "loss": 1.4035,
      "step": 5356
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2216216216216217e-05,
      "loss": 1.4908,
      "step": 5357
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2214735283228433e-05,
      "loss": 1.081,
      "step": 5358
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2213254350240652e-05,
      "loss": 1.2205,
      "step": 5359
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.221177341725287e-05,
      "loss": 1.3323,
      "step": 5360
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2210292484265087e-05,
      "loss": 1.3274,
      "step": 5361
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2208811551277307e-05,
      "loss": 1.4316,
      "step": 5362
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2207330618289524e-05,
      "loss": 1.5358,
      "step": 5363
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.220584968530174e-05,
      "loss": 1.5152,
      "step": 5364
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.220436875231396e-05,
      "loss": 1.2555,
      "step": 5365
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2202887819326177e-05,
      "loss": 1.3705,
      "step": 5366
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2201406886338394e-05,
      "loss": 1.3449,
      "step": 5367
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.219992595335061e-05,
      "loss": 1.2549,
      "step": 5368
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2198445020362831e-05,
      "loss": 1.3795,
      "step": 5369
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2196964087375047e-05,
      "loss": 1.2464,
      "step": 5370
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2195483154387264e-05,
      "loss": 1.4726,
      "step": 5371
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2194002221399484e-05,
      "loss": 1.4287,
      "step": 5372
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.2192521288411701e-05,
      "loss": 1.2682,
      "step": 5373
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2191040355423917e-05,
      "loss": 1.4727,
      "step": 5374
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2189559422436134e-05,
      "loss": 1.1484,
      "step": 5375
    },
    {
      "epoch": 1.19,
      "eval_loss": NaN,
      "eval_runtime": 4.3714,
      "eval_samples_per_second": 353.205,
      "eval_steps_per_second": 44.151,
      "step": 5375
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2188078489448354e-05,
      "loss": 1.2845,
      "step": 5376
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2186597556460571e-05,
      "loss": 1.2327,
      "step": 5377
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2185116623472789e-05,
      "loss": 1.2111,
      "step": 5378
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2183635690485008e-05,
      "loss": 1.2014,
      "step": 5379
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2182154757497224e-05,
      "loss": 1.4499,
      "step": 5380
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2180673824509441e-05,
      "loss": 1.2874,
      "step": 5381
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.217919289152166e-05,
      "loss": 1.3028,
      "step": 5382
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2177711958533878e-05,
      "loss": 1.4208,
      "step": 5383
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2176231025546095e-05,
      "loss": 1.4875,
      "step": 5384
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2174750092558311e-05,
      "loss": 1.3173,
      "step": 5385
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.217326915957053e-05,
      "loss": 1.3678,
      "step": 5386
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2171788226582748e-05,
      "loss": 1.0723,
      "step": 5387
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2170307293594966e-05,
      "loss": 1.4767,
      "step": 5388
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2168826360607185e-05,
      "loss": 1.2579,
      "step": 5389
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.21673454276194e-05,
      "loss": 1.5004,
      "step": 5390
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2165864494631618e-05,
      "loss": 1.3817,
      "step": 5391
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2164383561643837e-05,
      "loss": 1.2928,
      "step": 5392
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2162902628656055e-05,
      "loss": 1.3939,
      "step": 5393
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2161421695668272e-05,
      "loss": 1.2378,
      "step": 5394
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2159940762680488e-05,
      "loss": 1.1725,
      "step": 5395
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2158459829692707e-05,
      "loss": 1.2636,
      "step": 5396
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2156978896704925e-05,
      "loss": 1.5189,
      "step": 5397
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2155497963717142e-05,
      "loss": 1.3747,
      "step": 5398
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2154017030729362e-05,
      "loss": 1.2662,
      "step": 5399
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2152536097741579e-05,
      "loss": 1.3342,
      "step": 5400
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2151055164753795e-05,
      "loss": 1.2191,
      "step": 5401
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2149574231766012e-05,
      "loss": 1.1589,
      "step": 5402
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2148093298778232e-05,
      "loss": 1.0972,
      "step": 5403
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.214661236579045e-05,
      "loss": 1.3385,
      "step": 5404
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2145131432802667e-05,
      "loss": 1.3663,
      "step": 5405
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2143650499814886e-05,
      "loss": 1.1644,
      "step": 5406
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2142169566827102e-05,
      "loss": 1.3904,
      "step": 5407
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.214068863383932e-05,
      "loss": 1.0315,
      "step": 5408
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2139207700851538e-05,
      "loss": 1.2666,
      "step": 5409
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2137726767863756e-05,
      "loss": 1.156,
      "step": 5410
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2136245834875972e-05,
      "loss": 1.3762,
      "step": 5411
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.213476490188819e-05,
      "loss": 1.3897,
      "step": 5412
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2133283968900409e-05,
      "loss": 1.5372,
      "step": 5413
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2131803035912626e-05,
      "loss": 1.2983,
      "step": 5414
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2130322102924844e-05,
      "loss": 1.4455,
      "step": 5415
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2128841169937063e-05,
      "loss": 1.2483,
      "step": 5416
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2127360236949279e-05,
      "loss": 1.0905,
      "step": 5417
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2125879303961496e-05,
      "loss": 1.4774,
      "step": 5418
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2124398370973714e-05,
      "loss": 1.2707,
      "step": 5419
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2122917437985933e-05,
      "loss": 1.5863,
      "step": 5420
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.212143650499815e-05,
      "loss": 1.2716,
      "step": 5421
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2119955572010366e-05,
      "loss": 1.2429,
      "step": 5422
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2118474639022585e-05,
      "loss": 1.4202,
      "step": 5423
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2116993706034803e-05,
      "loss": 1.2598,
      "step": 5424
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.211551277304702e-05,
      "loss": 1.1833,
      "step": 5425
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.211403184005924e-05,
      "loss": 1.3854,
      "step": 5426
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2112550907071457e-05,
      "loss": 1.3591,
      "step": 5427
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2111069974083673e-05,
      "loss": 1.2624,
      "step": 5428
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.210958904109589e-05,
      "loss": 1.1236,
      "step": 5429
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.210810810810811e-05,
      "loss": 1.5436,
      "step": 5430
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2106627175120327e-05,
      "loss": 1.3491,
      "step": 5431
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2105146242132543e-05,
      "loss": 1.3694,
      "step": 5432
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2103665309144764e-05,
      "loss": 1.2596,
      "step": 5433
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.210218437615698e-05,
      "loss": 1.3241,
      "step": 5434
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2100703443169197e-05,
      "loss": 0.9655,
      "step": 5435
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2099222510181415e-05,
      "loss": 1.4405,
      "step": 5436
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2097741577193634e-05,
      "loss": 1.2172,
      "step": 5437
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.209626064420585e-05,
      "loss": 1.2825,
      "step": 5438
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2094779711218067e-05,
      "loss": 1.2125,
      "step": 5439
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2093298778230287e-05,
      "loss": 1.3289,
      "step": 5440
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2091817845242504e-05,
      "loss": 1.3488,
      "step": 5441
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2090336912254722e-05,
      "loss": 1.3878,
      "step": 5442
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.208885597926694e-05,
      "loss": 1.1378,
      "step": 5443
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2087375046279157e-05,
      "loss": 1.3911,
      "step": 5444
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2085894113291374e-05,
      "loss": 1.302,
      "step": 5445
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2084413180303592e-05,
      "loss": 1.2203,
      "step": 5446
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2082932247315811e-05,
      "loss": 1.329,
      "step": 5447
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2081451314328028e-05,
      "loss": 1.2,
      "step": 5448
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2079970381340244e-05,
      "loss": 1.4168,
      "step": 5449
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2078489448352463e-05,
      "loss": 1.3997,
      "step": 5450
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2077008515364681e-05,
      "loss": 1.4587,
      "step": 5451
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2075527582376898e-05,
      "loss": 1.2738,
      "step": 5452
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2074046649389114e-05,
      "loss": 1.1986,
      "step": 5453
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2072565716401334e-05,
      "loss": 1.1774,
      "step": 5454
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2071084783413551e-05,
      "loss": 1.3817,
      "step": 5455
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2069603850425769e-05,
      "loss": 1.2777,
      "step": 5456
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2068122917437988e-05,
      "loss": 1.2401,
      "step": 5457
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2066641984450205e-05,
      "loss": 1.1244,
      "step": 5458
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2065161051462421e-05,
      "loss": 1.1817,
      "step": 5459
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.206368011847464e-05,
      "loss": 1.3511,
      "step": 5460
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2062199185486858e-05,
      "loss": 1.4539,
      "step": 5461
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2060718252499075e-05,
      "loss": 1.5603,
      "step": 5462
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2059237319511293e-05,
      "loss": 1.4273,
      "step": 5463
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.2057756386523512e-05,
      "loss": 1.1862,
      "step": 5464
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2056275453535728e-05,
      "loss": 1.3885,
      "step": 5465
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2054794520547945e-05,
      "loss": 1.3037,
      "step": 5466
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2053313587560165e-05,
      "loss": 1.4923,
      "step": 5467
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2051832654572382e-05,
      "loss": 1.559,
      "step": 5468
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.20503517215846e-05,
      "loss": 1.2979,
      "step": 5469
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2048870788596815e-05,
      "loss": 1.3166,
      "step": 5470
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2047389855609035e-05,
      "loss": 1.301,
      "step": 5471
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2045908922621252e-05,
      "loss": 1.2647,
      "step": 5472
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.204442798963347e-05,
      "loss": 1.3611,
      "step": 5473
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2042947056645689e-05,
      "loss": 1.3763,
      "step": 5474
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2041466123657905e-05,
      "loss": 1.3138,
      "step": 5475
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2039985190670122e-05,
      "loss": 1.2375,
      "step": 5476
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2038504257682341e-05,
      "loss": 1.4669,
      "step": 5477
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2037023324694559e-05,
      "loss": 1.1647,
      "step": 5478
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2035542391706777e-05,
      "loss": 1.1552,
      "step": 5479
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2034061458718992e-05,
      "loss": 1.351,
      "step": 5480
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2032580525731212e-05,
      "loss": 1.1996,
      "step": 5481
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2031099592743429e-05,
      "loss": 1.2867,
      "step": 5482
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2029618659755647e-05,
      "loss": 1.0808,
      "step": 5483
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2028137726767866e-05,
      "loss": 1.3347,
      "step": 5484
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2026656793780083e-05,
      "loss": 1.5431,
      "step": 5485
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2025175860792299e-05,
      "loss": 1.0891,
      "step": 5486
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2023694927804517e-05,
      "loss": 1.3458,
      "step": 5487
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2022213994816736e-05,
      "loss": 1.3492,
      "step": 5488
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2020733061828953e-05,
      "loss": 1.3035,
      "step": 5489
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2019252128841171e-05,
      "loss": 1.3215,
      "step": 5490
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.201777119585339e-05,
      "loss": 1.2559,
      "step": 5491
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2016290262865606e-05,
      "loss": 1.4643,
      "step": 5492
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2014809329877823e-05,
      "loss": 1.3574,
      "step": 5493
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2013328396890043e-05,
      "loss": 1.3843,
      "step": 5494
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.201184746390226e-05,
      "loss": 1.4026,
      "step": 5495
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2010366530914476e-05,
      "loss": 1.4246,
      "step": 5496
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2008885597926694e-05,
      "loss": 1.5607,
      "step": 5497
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2007404664938913e-05,
      "loss": 1.3948,
      "step": 5498
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.200592373195113e-05,
      "loss": 1.188,
      "step": 5499
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2004442798963348e-05,
      "loss": 1.3777,
      "step": 5500
    },
    {
      "epoch": 1.21,
      "eval_loss": NaN,
      "eval_runtime": 4.3757,
      "eval_samples_per_second": 352.858,
      "eval_steps_per_second": 44.107,
      "step": 5500
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2002961865975567e-05,
      "loss": 1.3398,
      "step": 5501
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2001480932987783e-05,
      "loss": 1.363,
      "step": 5502
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.2e-05,
      "loss": 1.0738,
      "step": 5503
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.1998519067012218e-05,
      "loss": 1.1988,
      "step": 5504
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.1997038134024437e-05,
      "loss": 1.3809,
      "step": 5505
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.1995557201036655e-05,
      "loss": 1.3099,
      "step": 5506
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.199407626804887e-05,
      "loss": 1.3427,
      "step": 5507
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.199259533506109e-05,
      "loss": 1.2959,
      "step": 5508
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.1991114402073307e-05,
      "loss": 1.2035,
      "step": 5509
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.1989633469085525e-05,
      "loss": 1.5343,
      "step": 5510
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1988152536097744e-05,
      "loss": 1.2157,
      "step": 5511
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1986671603109961e-05,
      "loss": 1.2142,
      "step": 5512
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1985190670122177e-05,
      "loss": 1.3528,
      "step": 5513
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1983709737134395e-05,
      "loss": 1.3534,
      "step": 5514
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1982228804146614e-05,
      "loss": 1.2991,
      "step": 5515
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1980747871158831e-05,
      "loss": 1.2099,
      "step": 5516
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1979266938171047e-05,
      "loss": 1.2995,
      "step": 5517
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1977786005183266e-05,
      "loss": 1.0829,
      "step": 5518
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1976305072195484e-05,
      "loss": 1.5193,
      "step": 5519
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1974824139207701e-05,
      "loss": 1.0843,
      "step": 5520
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.197334320621992e-05,
      "loss": 1.2038,
      "step": 5521
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1971862273232138e-05,
      "loss": 1.4589,
      "step": 5522
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1970381340244354e-05,
      "loss": 1.348,
      "step": 5523
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1968900407256572e-05,
      "loss": 1.1807,
      "step": 5524
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.196741947426879e-05,
      "loss": 1.1883,
      "step": 5525
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1965938541281008e-05,
      "loss": 1.3332,
      "step": 5526
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1964457608293226e-05,
      "loss": 1.6181,
      "step": 5527
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1962976675305445e-05,
      "loss": 1.338,
      "step": 5528
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.196149574231766e-05,
      "loss": 1.3137,
      "step": 5529
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1960014809329878e-05,
      "loss": 1.4789,
      "step": 5530
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1958533876342096e-05,
      "loss": 1.21,
      "step": 5531
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1957052943354315e-05,
      "loss": 1.2965,
      "step": 5532
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1955572010366533e-05,
      "loss": 1.2896,
      "step": 5533
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1954091077378748e-05,
      "loss": 1.3287,
      "step": 5534
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1952610144390968e-05,
      "loss": 1.2,
      "step": 5535
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1951129211403185e-05,
      "loss": 1.2047,
      "step": 5536
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1949648278415403e-05,
      "loss": 1.3172,
      "step": 5537
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1948167345427622e-05,
      "loss": 1.4686,
      "step": 5538
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1946686412439838e-05,
      "loss": 1.2335,
      "step": 5539
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1945205479452055e-05,
      "loss": 1.2614,
      "step": 5540
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1943724546464273e-05,
      "loss": 1.1948,
      "step": 5541
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1942243613476492e-05,
      "loss": 1.2789,
      "step": 5542
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.194076268048871e-05,
      "loss": 1.1019,
      "step": 5543
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1939281747500925e-05,
      "loss": 1.4691,
      "step": 5544
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1937800814513144e-05,
      "loss": 1.2434,
      "step": 5545
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1936319881525362e-05,
      "loss": 1.4069,
      "step": 5546
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.193483894853758e-05,
      "loss": 1.0427,
      "step": 5547
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1933358015549797e-05,
      "loss": 1.1692,
      "step": 5548
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1931877082562016e-05,
      "loss": 1.2308,
      "step": 5549
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1930396149574232e-05,
      "loss": 1.3049,
      "step": 5550
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.192891521658645e-05,
      "loss": 1.3118,
      "step": 5551
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1927434283598669e-05,
      "loss": 1.2269,
      "step": 5552
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1925953350610886e-05,
      "loss": 1.3118,
      "step": 5553
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1924472417623104e-05,
      "loss": 1.3374,
      "step": 5554
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.1922991484635323e-05,
      "loss": 1.3926,
      "step": 5555
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1921510551647539e-05,
      "loss": 1.606,
      "step": 5556
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1920029618659756e-05,
      "loss": 1.3266,
      "step": 5557
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1918548685671974e-05,
      "loss": 1.1573,
      "step": 5558
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1917067752684193e-05,
      "loss": 1.274,
      "step": 5559
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1915586819696409e-05,
      "loss": 1.4149,
      "step": 5560
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1914105886708626e-05,
      "loss": 1.4264,
      "step": 5561
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1912624953720846e-05,
      "loss": 1.3093,
      "step": 5562
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1911144020733063e-05,
      "loss": 1.4188,
      "step": 5563
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.190966308774528e-05,
      "loss": 1.038,
      "step": 5564
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1908182154757497e-05,
      "loss": 1.2554,
      "step": 5565
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1906701221769716e-05,
      "loss": 1.1965,
      "step": 5566
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1905220288781933e-05,
      "loss": 1.1609,
      "step": 5567
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.190373935579415e-05,
      "loss": 1.24,
      "step": 5568
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.190225842280637e-05,
      "loss": 1.2498,
      "step": 5569
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1900777489818587e-05,
      "loss": 1.1748,
      "step": 5570
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1899296556830803e-05,
      "loss": 1.2528,
      "step": 5571
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1897815623843023e-05,
      "loss": 1.4021,
      "step": 5572
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.189633469085524e-05,
      "loss": 1.4468,
      "step": 5573
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1894853757867458e-05,
      "loss": 1.4207,
      "step": 5574
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1893372824879675e-05,
      "loss": 1.5116,
      "step": 5575
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1891891891891894e-05,
      "loss": 1.3638,
      "step": 5576
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.189041095890411e-05,
      "loss": 1.2786,
      "step": 5577
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1888930025916328e-05,
      "loss": 1.4554,
      "step": 5578
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1887449092928547e-05,
      "loss": 1.5136,
      "step": 5579
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1885968159940764e-05,
      "loss": 1.1856,
      "step": 5580
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.188448722695298e-05,
      "loss": 1.4854,
      "step": 5581
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1883006293965198e-05,
      "loss": 1.505,
      "step": 5582
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1881525360977417e-05,
      "loss": 1.3836,
      "step": 5583
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1880044427989634e-05,
      "loss": 1.5782,
      "step": 5584
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1878563495001852e-05,
      "loss": 1.4476,
      "step": 5585
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1877082562014071e-05,
      "loss": 1.4488,
      "step": 5586
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1875601629026287e-05,
      "loss": 1.3972,
      "step": 5587
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1874120696038504e-05,
      "loss": 1.2904,
      "step": 5588
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1872639763050724e-05,
      "loss": 1.462,
      "step": 5589
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1871158830062941e-05,
      "loss": 1.0902,
      "step": 5590
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1869677897075159e-05,
      "loss": 1.362,
      "step": 5591
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1868196964087375e-05,
      "loss": 1.6401,
      "step": 5592
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1866716031099594e-05,
      "loss": 1.33,
      "step": 5593
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1865235098111811e-05,
      "loss": 1.3625,
      "step": 5594
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1863754165124029e-05,
      "loss": 1.1941,
      "step": 5595
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1862273232136248e-05,
      "loss": 1.1522,
      "step": 5596
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1860792299148466e-05,
      "loss": 1.5044,
      "step": 5597
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1859311366160681e-05,
      "loss": 1.1901,
      "step": 5598
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1857830433172899e-05,
      "loss": 1.5098,
      "step": 5599
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1856349500185118e-05,
      "loss": 1.4011,
      "step": 5600
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1854868567197336e-05,
      "loss": 1.3468,
      "step": 5601
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1853387634209551e-05,
      "loss": 1.3607,
      "step": 5602
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.185190670122177e-05,
      "loss": 1.1693,
      "step": 5603
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1850425768233988e-05,
      "loss": 1.3726,
      "step": 5604
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1848944835246206e-05,
      "loss": 1.1184,
      "step": 5605
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1847463902258425e-05,
      "loss": 1.426,
      "step": 5606
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1845982969270642e-05,
      "loss": 1.1998,
      "step": 5607
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1844502036282858e-05,
      "loss": 1.446,
      "step": 5608
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1843021103295076e-05,
      "loss": 1.2718,
      "step": 5609
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1841540170307295e-05,
      "loss": 1.0994,
      "step": 5610
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1840059237319512e-05,
      "loss": 1.3701,
      "step": 5611
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.183857830433173e-05,
      "loss": 1.2468,
      "step": 5612
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.183709737134395e-05,
      "loss": 1.1912,
      "step": 5613
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1835616438356165e-05,
      "loss": 1.3875,
      "step": 5614
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1834135505368383e-05,
      "loss": 1.1716,
      "step": 5615
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.18326545723806e-05,
      "loss": 1.383,
      "step": 5616
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.183117363939282e-05,
      "loss": 1.1629,
      "step": 5617
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1829692706405037e-05,
      "loss": 1.3998,
      "step": 5618
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1828211773417253e-05,
      "loss": 1.49,
      "step": 5619
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1826730840429472e-05,
      "loss": 1.2767,
      "step": 5620
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.182524990744169e-05,
      "loss": 1.1216,
      "step": 5621
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1823768974453907e-05,
      "loss": 1.4093,
      "step": 5622
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1822288041466126e-05,
      "loss": 1.4061,
      "step": 5623
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1820807108478342e-05,
      "loss": 1.5115,
      "step": 5624
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.181932617549056e-05,
      "loss": 1.4756,
      "step": 5625
    },
    {
      "epoch": 1.24,
      "eval_loss": NaN,
      "eval_runtime": 4.3664,
      "eval_samples_per_second": 353.607,
      "eval_steps_per_second": 44.201,
      "step": 5625
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1817845242502777e-05,
      "loss": 1.2431,
      "step": 5626
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1816364309514996e-05,
      "loss": 1.3656,
      "step": 5627
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1814883376527214e-05,
      "loss": 1.1752,
      "step": 5628
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.181340244353943e-05,
      "loss": 1.145,
      "step": 5629
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1811921510551649e-05,
      "loss": 1.3255,
      "step": 5630
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1810440577563866e-05,
      "loss": 1.2498,
      "step": 5631
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1808959644576084e-05,
      "loss": 1.3773,
      "step": 5632
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1807478711588301e-05,
      "loss": 1.4117,
      "step": 5633
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.180599777860052e-05,
      "loss": 1.2152,
      "step": 5634
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1804516845612736e-05,
      "loss": 1.3705,
      "step": 5635
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1803035912624954e-05,
      "loss": 1.4012,
      "step": 5636
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1801554979637173e-05,
      "loss": 1.5471,
      "step": 5637
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.180007404664939e-05,
      "loss": 1.0215,
      "step": 5638
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1798593113661608e-05,
      "loss": 1.2072,
      "step": 5639
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1797112180673827e-05,
      "loss": 1.1179,
      "step": 5640
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1795631247686043e-05,
      "loss": 1.2987,
      "step": 5641
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.179415031469826e-05,
      "loss": 1.3556,
      "step": 5642
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1792669381710478e-05,
      "loss": 1.4659,
      "step": 5643
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1791188448722697e-05,
      "loss": 1.1149,
      "step": 5644
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.1789707515734913e-05,
      "loss": 1.1386,
      "step": 5645
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.178822658274713e-05,
      "loss": 1.2108,
      "step": 5646
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.178674564975935e-05,
      "loss": 1.3366,
      "step": 5647
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1785264716771567e-05,
      "loss": 1.2812,
      "step": 5648
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1783783783783785e-05,
      "loss": 1.09,
      "step": 5649
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1782302850796e-05,
      "loss": 1.4682,
      "step": 5650
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.178082191780822e-05,
      "loss": 1.3368,
      "step": 5651
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1779340984820437e-05,
      "loss": 1.2195,
      "step": 5652
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1777860051832655e-05,
      "loss": 1.3808,
      "step": 5653
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1776379118844874e-05,
      "loss": 1.4918,
      "step": 5654
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1774898185857092e-05,
      "loss": 1.3476,
      "step": 5655
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1773417252869307e-05,
      "loss": 1.3515,
      "step": 5656
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1771936319881527e-05,
      "loss": 1.3882,
      "step": 5657
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1770455386893744e-05,
      "loss": 1.201,
      "step": 5658
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1768974453905962e-05,
      "loss": 1.4247,
      "step": 5659
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.176749352091818e-05,
      "loss": 1.3246,
      "step": 5660
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1766012587930398e-05,
      "loss": 1.3161,
      "step": 5661
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1764531654942614e-05,
      "loss": 1.3076,
      "step": 5662
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1763050721954832e-05,
      "loss": 1.4078,
      "step": 5663
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1761569788967051e-05,
      "loss": 0.9644,
      "step": 5664
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1760088855979269e-05,
      "loss": 1.3679,
      "step": 5665
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1758607922991484e-05,
      "loss": 1.1854,
      "step": 5666
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1757126990003704e-05,
      "loss": 1.2742,
      "step": 5667
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1755646057015921e-05,
      "loss": 1.4125,
      "step": 5668
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1754165124028139e-05,
      "loss": 1.3244,
      "step": 5669
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1752684191040356e-05,
      "loss": 1.2516,
      "step": 5670
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1751203258052575e-05,
      "loss": 1.2529,
      "step": 5671
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1749722325064791e-05,
      "loss": 1.3065,
      "step": 5672
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1748241392077009e-05,
      "loss": 1.3898,
      "step": 5673
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1746760459089228e-05,
      "loss": 1.2632,
      "step": 5674
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1745279526101445e-05,
      "loss": 1.2111,
      "step": 5675
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1743798593113663e-05,
      "loss": 1.2783,
      "step": 5676
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1742317660125879e-05,
      "loss": 1.375,
      "step": 5677
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1740836727138098e-05,
      "loss": 1.3588,
      "step": 5678
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1739355794150315e-05,
      "loss": 1.3347,
      "step": 5679
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1737874861162533e-05,
      "loss": 1.3145,
      "step": 5680
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1736393928174752e-05,
      "loss": 1.1681,
      "step": 5681
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.173491299518697e-05,
      "loss": 1.1243,
      "step": 5682
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1733432062199186e-05,
      "loss": 1.2637,
      "step": 5683
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1731951129211405e-05,
      "loss": 1.3049,
      "step": 5684
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1730470196223622e-05,
      "loss": 1.0941,
      "step": 5685
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.172898926323584e-05,
      "loss": 1.4168,
      "step": 5686
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1727508330248056e-05,
      "loss": 1.5123,
      "step": 5687
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1726027397260275e-05,
      "loss": 1.2223,
      "step": 5688
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1724546464272492e-05,
      "loss": 1.3794,
      "step": 5689
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.172306553128471e-05,
      "loss": 1.2275,
      "step": 5690
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1721584598296929e-05,
      "loss": 1.1932,
      "step": 5691
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1720103665309147e-05,
      "loss": 1.1963,
      "step": 5692
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1718622732321362e-05,
      "loss": 1.3243,
      "step": 5693
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.171714179933358e-05,
      "loss": 1.5616,
      "step": 5694
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1715660866345799e-05,
      "loss": 1.2545,
      "step": 5695
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1714179933358017e-05,
      "loss": 1.3769,
      "step": 5696
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1712699000370234e-05,
      "loss": 1.3163,
      "step": 5697
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1711218067382453e-05,
      "loss": 1.3518,
      "step": 5698
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.170973713439467e-05,
      "loss": 1.1845,
      "step": 5699
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1708256201406887e-05,
      "loss": 1.5408,
      "step": 5700
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1706775268419106e-05,
      "loss": 1.4608,
      "step": 5701
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1705294335431323e-05,
      "loss": 1.4363,
      "step": 5702
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1703813402443541e-05,
      "loss": 1.2035,
      "step": 5703
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1702332469455757e-05,
      "loss": 1.3613,
      "step": 5704
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1700851536467976e-05,
      "loss": 1.2298,
      "step": 5705
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1699370603480194e-05,
      "loss": 1.452,
      "step": 5706
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1697889670492411e-05,
      "loss": 1.5948,
      "step": 5707
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.169640873750463e-05,
      "loss": 1.326,
      "step": 5708
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1694927804516846e-05,
      "loss": 1.2787,
      "step": 5709
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1693446871529064e-05,
      "loss": 1.4056,
      "step": 5710
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1691965938541281e-05,
      "loss": 1.3006,
      "step": 5711
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.16904850055535e-05,
      "loss": 1.3491,
      "step": 5712
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1689004072565718e-05,
      "loss": 1.3836,
      "step": 5713
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1687523139577934e-05,
      "loss": 1.2349,
      "step": 5714
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1686042206590153e-05,
      "loss": 1.4217,
      "step": 5715
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.168456127360237e-05,
      "loss": 1.4413,
      "step": 5716
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1683080340614588e-05,
      "loss": 1.2919,
      "step": 5717
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1681599407626807e-05,
      "loss": 1.1468,
      "step": 5718
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1680118474639025e-05,
      "loss": 1.3842,
      "step": 5719
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.167863754165124e-05,
      "loss": 1.4924,
      "step": 5720
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1677156608663458e-05,
      "loss": 1.3402,
      "step": 5721
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1675675675675677e-05,
      "loss": 1.2454,
      "step": 5722
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1674194742687895e-05,
      "loss": 1.3305,
      "step": 5723
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1672713809700112e-05,
      "loss": 1.5121,
      "step": 5724
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1671232876712331e-05,
      "loss": 1.484,
      "step": 5725
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1669751943724547e-05,
      "loss": 1.3064,
      "step": 5726
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1668271010736765e-05,
      "loss": 1.3886,
      "step": 5727
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1666790077748982e-05,
      "loss": 1.458,
      "step": 5728
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1665309144761201e-05,
      "loss": 1.2822,
      "step": 5729
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1663828211773417e-05,
      "loss": 1.1248,
      "step": 5730
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1662347278785635e-05,
      "loss": 1.1363,
      "step": 5731
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1660866345797854e-05,
      "loss": 1.3461,
      "step": 5732
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1659385412810072e-05,
      "loss": 1.5072,
      "step": 5733
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1657904479822289e-05,
      "loss": 1.3526,
      "step": 5734
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1656423546834508e-05,
      "loss": 1.2139,
      "step": 5735
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.1654942613846724e-05,
      "loss": 1.3382,
      "step": 5736
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1653461680858942e-05,
      "loss": 1.4188,
      "step": 5737
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1651980747871159e-05,
      "loss": 1.2546,
      "step": 5738
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1650499814883378e-05,
      "loss": 1.337,
      "step": 5739
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1649018881895596e-05,
      "loss": 1.2206,
      "step": 5740
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1647537948907812e-05,
      "loss": 1.3006,
      "step": 5741
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1646057015920031e-05,
      "loss": 1.2873,
      "step": 5742
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1644576082932248e-05,
      "loss": 1.1572,
      "step": 5743
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1643095149944466e-05,
      "loss": 1.3165,
      "step": 5744
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1641614216956682e-05,
      "loss": 1.2702,
      "step": 5745
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1640133283968903e-05,
      "loss": 1.544,
      "step": 5746
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1638652350981118e-05,
      "loss": 1.3848,
      "step": 5747
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1637171417993336e-05,
      "loss": 1.3074,
      "step": 5748
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1635690485005555e-05,
      "loss": 1.2344,
      "step": 5749
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1634209552017773e-05,
      "loss": 1.1598,
      "step": 5750
    },
    {
      "epoch": 1.27,
      "eval_loss": NaN,
      "eval_runtime": 4.374,
      "eval_samples_per_second": 352.994,
      "eval_steps_per_second": 44.124,
      "step": 5750
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1632728619029989e-05,
      "loss": 1.2456,
      "step": 5751
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1631247686042208e-05,
      "loss": 1.4529,
      "step": 5752
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1629766753054425e-05,
      "loss": 1.3042,
      "step": 5753
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1628285820066643e-05,
      "loss": 1.2399,
      "step": 5754
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.162680488707886e-05,
      "loss": 1.2214,
      "step": 5755
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.162532395409108e-05,
      "loss": 1.2498,
      "step": 5756
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1623843021103295e-05,
      "loss": 1.4129,
      "step": 5757
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1622362088115513e-05,
      "loss": 1.1903,
      "step": 5758
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1620881155127732e-05,
      "loss": 1.1577,
      "step": 5759
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.161940022213995e-05,
      "loss": 1.4132,
      "step": 5760
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1617919289152167e-05,
      "loss": 1.3727,
      "step": 5761
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1616438356164383e-05,
      "loss": 1.1554,
      "step": 5762
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1614957423176602e-05,
      "loss": 1.221,
      "step": 5763
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.161347649018882e-05,
      "loss": 1.251,
      "step": 5764
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1611995557201037e-05,
      "loss": 1.1578,
      "step": 5765
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1610514624213256e-05,
      "loss": 1.4014,
      "step": 5766
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1609033691225474e-05,
      "loss": 1.2147,
      "step": 5767
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.160755275823769e-05,
      "loss": 1.1972,
      "step": 5768
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1606071825249909e-05,
      "loss": 1.3686,
      "step": 5769
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1604590892262126e-05,
      "loss": 1.4286,
      "step": 5770
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1603109959274344e-05,
      "loss": 1.0964,
      "step": 5771
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.160162902628656e-05,
      "loss": 1.5233,
      "step": 5772
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1600148093298779e-05,
      "loss": 1.4108,
      "step": 5773
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1598667160310997e-05,
      "loss": 1.5833,
      "step": 5774
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1597186227323214e-05,
      "loss": 1.4219,
      "step": 5775
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1595705294335433e-05,
      "loss": 1.5429,
      "step": 5776
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.159422436134765e-05,
      "loss": 1.3644,
      "step": 5777
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1592743428359867e-05,
      "loss": 1.3113,
      "step": 5778
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1591262495372084e-05,
      "loss": 1.1582,
      "step": 5779
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1589781562384303e-05,
      "loss": 1.3429,
      "step": 5780
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.158830062939652e-05,
      "loss": 1.4548,
      "step": 5781
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.1586819696408738e-05,
      "loss": 1.2565,
      "step": 5782
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1585338763420958e-05,
      "loss": 1.3109,
      "step": 5783
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1583857830433173e-05,
      "loss": 1.2762,
      "step": 5784
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1582376897445391e-05,
      "loss": 1.3572,
      "step": 5785
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.158089596445761e-05,
      "loss": 1.3571,
      "step": 5786
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1579415031469828e-05,
      "loss": 1.1748,
      "step": 5787
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1577934098482045e-05,
      "loss": 1.5468,
      "step": 5788
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1576453165494261e-05,
      "loss": 1.3871,
      "step": 5789
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.157497223250648e-05,
      "loss": 1.2463,
      "step": 5790
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1573491299518698e-05,
      "loss": 1.2469,
      "step": 5791
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1572010366530915e-05,
      "loss": 1.2004,
      "step": 5792
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1570529433543134e-05,
      "loss": 1.3661,
      "step": 5793
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.156904850055535e-05,
      "loss": 1.2223,
      "step": 5794
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1567567567567568e-05,
      "loss": 1.3331,
      "step": 5795
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1566086634579787e-05,
      "loss": 1.1821,
      "step": 5796
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1564605701592004e-05,
      "loss": 1.4639,
      "step": 5797
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1563124768604222e-05,
      "loss": 1.3531,
      "step": 5798
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1561643835616438e-05,
      "loss": 1.2142,
      "step": 5799
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1560162902628657e-05,
      "loss": 1.3571,
      "step": 5800
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1558681969640875e-05,
      "loss": 1.5619,
      "step": 5801
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1557201036653092e-05,
      "loss": 1.3143,
      "step": 5802
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1555720103665311e-05,
      "loss": 1.5419,
      "step": 5803
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1554239170677529e-05,
      "loss": 1.4906,
      "step": 5804
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1552758237689745e-05,
      "loss": 1.4968,
      "step": 5805
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1551277304701962e-05,
      "loss": 1.4504,
      "step": 5806
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1549796371714181e-05,
      "loss": 1.2522,
      "step": 5807
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1548315438726399e-05,
      "loss": 1.372,
      "step": 5808
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1546834505738615e-05,
      "loss": 1.1283,
      "step": 5809
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1545353572750836e-05,
      "loss": 1.3112,
      "step": 5810
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1543872639763051e-05,
      "loss": 1.3335,
      "step": 5811
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1542391706775269e-05,
      "loss": 1.4134,
      "step": 5812
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1540910773787488e-05,
      "loss": 1.3767,
      "step": 5813
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1539429840799706e-05,
      "loss": 1.2453,
      "step": 5814
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1537948907811921e-05,
      "loss": 1.2156,
      "step": 5815
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1536467974824139e-05,
      "loss": 1.4223,
      "step": 5816
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1534987041836358e-05,
      "loss": 1.7047,
      "step": 5817
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1533506108848576e-05,
      "loss": 1.1201,
      "step": 5818
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1532025175860793e-05,
      "loss": 1.329,
      "step": 5819
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1530544242873012e-05,
      "loss": 1.453,
      "step": 5820
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1529063309885228e-05,
      "loss": 1.366,
      "step": 5821
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1527582376897446e-05,
      "loss": 1.2824,
      "step": 5822
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1526101443909663e-05,
      "loss": 1.4663,
      "step": 5823
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1524620510921883e-05,
      "loss": 1.3772,
      "step": 5824
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.15231395779341e-05,
      "loss": 1.1763,
      "step": 5825
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1521658644946316e-05,
      "loss": 1.3034,
      "step": 5826
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1520177711958535e-05,
      "loss": 1.4683,
      "step": 5827
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1518696778970753e-05,
      "loss": 0.9726,
      "step": 5828
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.151721584598297e-05,
      "loss": 1.4877,
      "step": 5829
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.151573491299519e-05,
      "loss": 1.4,
      "step": 5830
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1514253980007407e-05,
      "loss": 1.2181,
      "step": 5831
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1512773047019623e-05,
      "loss": 1.109,
      "step": 5832
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.151129211403184e-05,
      "loss": 1.2119,
      "step": 5833
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.150981118104406e-05,
      "loss": 1.3619,
      "step": 5834
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1508330248056277e-05,
      "loss": 1.1049,
      "step": 5835
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1506849315068493e-05,
      "loss": 1.5315,
      "step": 5836
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1505368382080712e-05,
      "loss": 1.3395,
      "step": 5837
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.150388744909293e-05,
      "loss": 1.2954,
      "step": 5838
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1502406516105147e-05,
      "loss": 1.2261,
      "step": 5839
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1500925583117364e-05,
      "loss": 1.383,
      "step": 5840
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1499444650129584e-05,
      "loss": 1.3407,
      "step": 5841
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.14979637171418e-05,
      "loss": 1.4378,
      "step": 5842
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1496482784154017e-05,
      "loss": 1.3228,
      "step": 5843
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1495001851166236e-05,
      "loss": 1.2504,
      "step": 5844
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1493520918178454e-05,
      "loss": 1.4051,
      "step": 5845
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1492039985190671e-05,
      "loss": 1.2871,
      "step": 5846
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.149055905220289e-05,
      "loss": 1.1391,
      "step": 5847
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1489078119215106e-05,
      "loss": 1.4871,
      "step": 5848
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1487597186227324e-05,
      "loss": 1.295,
      "step": 5849
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1486116253239541e-05,
      "loss": 1.3729,
      "step": 5850
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.148463532025176e-05,
      "loss": 1.3436,
      "step": 5851
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1483154387263978e-05,
      "loss": 1.1469,
      "step": 5852
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1481673454276194e-05,
      "loss": 1.2425,
      "step": 5853
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1480192521288413e-05,
      "loss": 1.1925,
      "step": 5854
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.147871158830063e-05,
      "loss": 1.3724,
      "step": 5855
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1477230655312848e-05,
      "loss": 0.9823,
      "step": 5856
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1475749722325064e-05,
      "loss": 1.2442,
      "step": 5857
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1474268789337283e-05,
      "loss": 1.4639,
      "step": 5858
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.14727878563495e-05,
      "loss": 1.4555,
      "step": 5859
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1471306923361718e-05,
      "loss": 1.3287,
      "step": 5860
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1469825990373937e-05,
      "loss": 1.2063,
      "step": 5861
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1468345057386155e-05,
      "loss": 1.2072,
      "step": 5862
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.146686412439837e-05,
      "loss": 1.1185,
      "step": 5863
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.146538319141059e-05,
      "loss": 1.2301,
      "step": 5864
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1463902258422807e-05,
      "loss": 1.1205,
      "step": 5865
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1462421325435025e-05,
      "loss": 1.215,
      "step": 5866
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1460940392447243e-05,
      "loss": 1.3839,
      "step": 5867
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1459459459459462e-05,
      "loss": 1.3629,
      "step": 5868
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1457978526471678e-05,
      "loss": 1.3484,
      "step": 5869
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1456497593483895e-05,
      "loss": 1.4601,
      "step": 5870
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1455016660496114e-05,
      "loss": 1.2809,
      "step": 5871
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.1453535727508332e-05,
      "loss": 1.1719,
      "step": 5872
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1452054794520548e-05,
      "loss": 1.2549,
      "step": 5873
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1450573861532765e-05,
      "loss": 1.2937,
      "step": 5874
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1449092928544984e-05,
      "loss": 1.0755,
      "step": 5875
    },
    {
      "epoch": 1.3,
      "eval_loss": NaN,
      "eval_runtime": 4.3786,
      "eval_samples_per_second": 352.626,
      "eval_steps_per_second": 44.078,
      "step": 5875
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1447611995557202e-05,
      "loss": 1.1133,
      "step": 5876
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.144613106256942e-05,
      "loss": 1.4457,
      "step": 5877
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1444650129581639e-05,
      "loss": 1.3184,
      "step": 5878
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1443169196593854e-05,
      "loss": 1.4779,
      "step": 5879
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1441688263606072e-05,
      "loss": 1.6904,
      "step": 5880
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1440207330618291e-05,
      "loss": 1.4495,
      "step": 5881
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1438726397630509e-05,
      "loss": 1.1902,
      "step": 5882
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1437245464642726e-05,
      "loss": 1.1605,
      "step": 5883
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1435764531654942e-05,
      "loss": 1.1197,
      "step": 5884
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1434283598667161e-05,
      "loss": 1.1876,
      "step": 5885
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1432802665679379e-05,
      "loss": 1.2575,
      "step": 5886
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1431321732691596e-05,
      "loss": 1.2059,
      "step": 5887
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1429840799703815e-05,
      "loss": 1.1553,
      "step": 5888
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1428359866716033e-05,
      "loss": 1.0549,
      "step": 5889
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1426878933728249e-05,
      "loss": 1.2679,
      "step": 5890
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1425398000740466e-05,
      "loss": 1.1513,
      "step": 5891
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1423917067752686e-05,
      "loss": 1.2301,
      "step": 5892
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1422436134764903e-05,
      "loss": 1.3219,
      "step": 5893
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1420955201777119e-05,
      "loss": 1.2227,
      "step": 5894
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.141947426878934e-05,
      "loss": 1.2448,
      "step": 5895
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1417993335801556e-05,
      "loss": 1.2063,
      "step": 5896
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1416512402813773e-05,
      "loss": 1.3412,
      "step": 5897
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1415031469825992e-05,
      "loss": 1.2084,
      "step": 5898
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.141355053683821e-05,
      "loss": 1.2847,
      "step": 5899
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1412069603850426e-05,
      "loss": 1.2693,
      "step": 5900
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1410588670862643e-05,
      "loss": 1.3498,
      "step": 5901
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1409107737874862e-05,
      "loss": 1.1841,
      "step": 5902
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.140762680488708e-05,
      "loss": 1.2664,
      "step": 5903
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1406145871899297e-05,
      "loss": 1.3079,
      "step": 5904
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1404664938911517e-05,
      "loss": 1.357,
      "step": 5905
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1403184005923732e-05,
      "loss": 1.1937,
      "step": 5906
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.140170307293595e-05,
      "loss": 1.2849,
      "step": 5907
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1400222139948167e-05,
      "loss": 1.1564,
      "step": 5908
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1398741206960387e-05,
      "loss": 1.5435,
      "step": 5909
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1397260273972604e-05,
      "loss": 1.2136,
      "step": 5910
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.139577934098482e-05,
      "loss": 1.5206,
      "step": 5911
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.139429840799704e-05,
      "loss": 1.254,
      "step": 5912
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1392817475009257e-05,
      "loss": 1.3394,
      "step": 5913
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1391336542021474e-05,
      "loss": 1.3999,
      "step": 5914
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1389855609033693e-05,
      "loss": 1.4589,
      "step": 5915
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1388374676045911e-05,
      "loss": 1.2603,
      "step": 5916
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1386893743058127e-05,
      "loss": 1.3638,
      "step": 5917
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.1385412810070344e-05,
      "loss": 1.3073,
      "step": 5918
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1383931877082564e-05,
      "loss": 1.6126,
      "step": 5919
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1382450944094781e-05,
      "loss": 1.3523,
      "step": 5920
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1380970011106997e-05,
      "loss": 1.3929,
      "step": 5921
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1379489078119216e-05,
      "loss": 1.2798,
      "step": 5922
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1378008145131434e-05,
      "loss": 1.5304,
      "step": 5923
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1376527212143651e-05,
      "loss": 1.1088,
      "step": 5924
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.137504627915587e-05,
      "loss": 1.2932,
      "step": 5925
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1373565346168088e-05,
      "loss": 1.1361,
      "step": 5926
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1372084413180304e-05,
      "loss": 1.5346,
      "step": 5927
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1370603480192521e-05,
      "loss": 1.0766,
      "step": 5928
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.136912254720474e-05,
      "loss": 1.287,
      "step": 5929
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1367641614216958e-05,
      "loss": 1.2009,
      "step": 5930
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1366160681229175e-05,
      "loss": 1.432,
      "step": 5931
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1364679748241395e-05,
      "loss": 1.2819,
      "step": 5932
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.136319881525361e-05,
      "loss": 1.2251,
      "step": 5933
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1361717882265828e-05,
      "loss": 1.4804,
      "step": 5934
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1360236949278046e-05,
      "loss": 1.3764,
      "step": 5935
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1358756016290265e-05,
      "loss": 1.3699,
      "step": 5936
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.135727508330248e-05,
      "loss": 1.3572,
      "step": 5937
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1355794150314698e-05,
      "loss": 1.2283,
      "step": 5938
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1354313217326917e-05,
      "loss": 1.3389,
      "step": 5939
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1352832284339135e-05,
      "loss": 1.4635,
      "step": 5940
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1351351351351352e-05,
      "loss": 1.2384,
      "step": 5941
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1349870418363572e-05,
      "loss": 1.4065,
      "step": 5942
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1348389485375787e-05,
      "loss": 1.4081,
      "step": 5943
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1346908552388005e-05,
      "loss": 1.3763,
      "step": 5944
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1345427619400222e-05,
      "loss": 1.4942,
      "step": 5945
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1343946686412442e-05,
      "loss": 1.1083,
      "step": 5946
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1342465753424659e-05,
      "loss": 1.2015,
      "step": 5947
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1340984820436875e-05,
      "loss": 1.5063,
      "step": 5948
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1339503887449094e-05,
      "loss": 1.3123,
      "step": 5949
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1338022954461312e-05,
      "loss": 1.2212,
      "step": 5950
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.133654202147353e-05,
      "loss": 1.5034,
      "step": 5951
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1335061088485747e-05,
      "loss": 1.2045,
      "step": 5952
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1333580155497966e-05,
      "loss": 1.2155,
      "step": 5953
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1332099222510182e-05,
      "loss": 1.4204,
      "step": 5954
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.13306182895224e-05,
      "loss": 1.3833,
      "step": 5955
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1329137356534618e-05,
      "loss": 1.1049,
      "step": 5956
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1327656423546836e-05,
      "loss": 1.3952,
      "step": 5957
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1326175490559052e-05,
      "loss": 1.2094,
      "step": 5958
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1324694557571273e-05,
      "loss": 1.1512,
      "step": 5959
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1323213624583489e-05,
      "loss": 1.3276,
      "step": 5960
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1321732691595706e-05,
      "loss": 1.5111,
      "step": 5961
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1320251758607924e-05,
      "loss": 1.3233,
      "step": 5962
    },
    {
      "epoch": 1.31,
      "learning_rate": 1.1318770825620143e-05,
      "loss": 1.3552,
      "step": 5963
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1317289892632359e-05,
      "loss": 1.1383,
      "step": 5964
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1315808959644576e-05,
      "loss": 1.2759,
      "step": 5965
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1314328026656795e-05,
      "loss": 1.4141,
      "step": 5966
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1312847093669013e-05,
      "loss": 1.2242,
      "step": 5967
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.131136616068123e-05,
      "loss": 1.3222,
      "step": 5968
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1309885227693446e-05,
      "loss": 1.38,
      "step": 5969
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1308404294705665e-05,
      "loss": 1.3657,
      "step": 5970
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1306923361717883e-05,
      "loss": 1.4045,
      "step": 5971
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.13054424287301e-05,
      "loss": 1.5011,
      "step": 5972
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.130396149574232e-05,
      "loss": 1.1363,
      "step": 5973
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1302480562754537e-05,
      "loss": 1.1741,
      "step": 5974
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1300999629766753e-05,
      "loss": 1.2814,
      "step": 5975
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1299518696778972e-05,
      "loss": 1.3775,
      "step": 5976
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.129803776379119e-05,
      "loss": 1.2296,
      "step": 5977
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1296556830803407e-05,
      "loss": 1.2938,
      "step": 5978
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1295075897815623e-05,
      "loss": 1.2616,
      "step": 5979
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1293594964827844e-05,
      "loss": 1.3961,
      "step": 5980
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.129211403184006e-05,
      "loss": 1.2625,
      "step": 5981
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1290633098852277e-05,
      "loss": 1.013,
      "step": 5982
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1289152165864497e-05,
      "loss": 1.4476,
      "step": 5983
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1287671232876714e-05,
      "loss": 1.2809,
      "step": 5984
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.128619029988893e-05,
      "loss": 1.3004,
      "step": 5985
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1284709366901147e-05,
      "loss": 1.1574,
      "step": 5986
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1283228433913367e-05,
      "loss": 1.1695,
      "step": 5987
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1281747500925584e-05,
      "loss": 1.3444,
      "step": 5988
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1280266567937802e-05,
      "loss": 1.3566,
      "step": 5989
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.127878563495002e-05,
      "loss": 1.2444,
      "step": 5990
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1277304701962237e-05,
      "loss": 1.3388,
      "step": 5991
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1275823768974454e-05,
      "loss": 1.4351,
      "step": 5992
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1274342835986673e-05,
      "loss": 1.3727,
      "step": 5993
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1272861902998891e-05,
      "loss": 1.4427,
      "step": 5994
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1271380970011108e-05,
      "loss": 1.4385,
      "step": 5995
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1269900037023324e-05,
      "loss": 1.2122,
      "step": 5996
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1268419104035543e-05,
      "loss": 1.3633,
      "step": 5997
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1266938171047761e-05,
      "loss": 1.2399,
      "step": 5998
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1265457238059978e-05,
      "loss": 1.3128,
      "step": 5999
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1263976305072198e-05,
      "loss": 1.3644,
      "step": 6000
    },
    {
      "epoch": 1.32,
      "eval_loss": NaN,
      "eval_runtime": 4.3767,
      "eval_samples_per_second": 352.779,
      "eval_steps_per_second": 44.097,
      "step": 6000
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1262495372084414e-05,
      "loss": 1.4355,
      "step": 6001
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1261014439096631e-05,
      "loss": 0.9599,
      "step": 6002
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1259533506108849e-05,
      "loss": 1.1065,
      "step": 6003
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1258052573121068e-05,
      "loss": 1.1648,
      "step": 6004
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1256571640133285e-05,
      "loss": 1.4026,
      "step": 6005
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1255090707145501e-05,
      "loss": 1.2956,
      "step": 6006
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.125360977415772e-05,
      "loss": 1.2641,
      "step": 6007
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.1252128841169938e-05,
      "loss": 1.1342,
      "step": 6008
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1250647908182155e-05,
      "loss": 1.2027,
      "step": 6009
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1249166975194375e-05,
      "loss": 1.2244,
      "step": 6010
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1247686042206592e-05,
      "loss": 1.5249,
      "step": 6011
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1246205109218808e-05,
      "loss": 1.2589,
      "step": 6012
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1244724176231025e-05,
      "loss": 1.2984,
      "step": 6013
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1243243243243245e-05,
      "loss": 1.2721,
      "step": 6014
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1241762310255462e-05,
      "loss": 1.2599,
      "step": 6015
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.124028137726768e-05,
      "loss": 1.2954,
      "step": 6016
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1238800444279899e-05,
      "loss": 1.4236,
      "step": 6017
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1237319511292115e-05,
      "loss": 1.1338,
      "step": 6018
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1235838578304332e-05,
      "loss": 1.2912,
      "step": 6019
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.123435764531655e-05,
      "loss": 1.1734,
      "step": 6020
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1232876712328769e-05,
      "loss": 1.3133,
      "step": 6021
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1231395779340985e-05,
      "loss": 1.3209,
      "step": 6022
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1229914846353202e-05,
      "loss": 1.311,
      "step": 6023
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1228433913365421e-05,
      "loss": 1.2217,
      "step": 6024
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1226952980377639e-05,
      "loss": 1.4262,
      "step": 6025
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1225472047389857e-05,
      "loss": 1.3959,
      "step": 6026
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1223991114402076e-05,
      "loss": 1.1906,
      "step": 6027
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1222510181414292e-05,
      "loss": 1.304,
      "step": 6028
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1221029248426509e-05,
      "loss": 1.3915,
      "step": 6029
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1219548315438727e-05,
      "loss": 1.4371,
      "step": 6030
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1218067382450946e-05,
      "loss": 1.2243,
      "step": 6031
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1216586449463163e-05,
      "loss": 1.2404,
      "step": 6032
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1215105516475379e-05,
      "loss": 1.6725,
      "step": 6033
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1213624583487598e-05,
      "loss": 1.3298,
      "step": 6034
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1212143650499816e-05,
      "loss": 1.1923,
      "step": 6035
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1210662717512033e-05,
      "loss": 1.5882,
      "step": 6036
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1209181784524251e-05,
      "loss": 1.2869,
      "step": 6037
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.120770085153647e-05,
      "loss": 1.2674,
      "step": 6038
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1206219918548686e-05,
      "loss": 1.2168,
      "step": 6039
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1204738985560903e-05,
      "loss": 1.1001,
      "step": 6040
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1203258052573123e-05,
      "loss": 1.4199,
      "step": 6041
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.120177711958534e-05,
      "loss": 1.4255,
      "step": 6042
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1200296186597556e-05,
      "loss": 1.5172,
      "step": 6043
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1198815253609777e-05,
      "loss": 1.2254,
      "step": 6044
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1197334320621993e-05,
      "loss": 1.3916,
      "step": 6045
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.119585338763421e-05,
      "loss": 1.2676,
      "step": 6046
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1194372454646428e-05,
      "loss": 1.1609,
      "step": 6047
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1192891521658647e-05,
      "loss": 1.3555,
      "step": 6048
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1191410588670863e-05,
      "loss": 1.2813,
      "step": 6049
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.118992965568308e-05,
      "loss": 1.1722,
      "step": 6050
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.11884487226953e-05,
      "loss": 1.2075,
      "step": 6051
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1186967789707517e-05,
      "loss": 1.3859,
      "step": 6052
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1185486856719735e-05,
      "loss": 1.0959,
      "step": 6053
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.1184005923731954e-05,
      "loss": 1.3818,
      "step": 6054
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.118252499074417e-05,
      "loss": 1.6045,
      "step": 6055
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1181044057756387e-05,
      "loss": 1.549,
      "step": 6056
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1179563124768605e-05,
      "loss": 1.3087,
      "step": 6057
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1178082191780824e-05,
      "loss": 1.2863,
      "step": 6058
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1176601258793041e-05,
      "loss": 1.3141,
      "step": 6059
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1175120325805257e-05,
      "loss": 1.2096,
      "step": 6060
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1173639392817476e-05,
      "loss": 1.3798,
      "step": 6061
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1172158459829694e-05,
      "loss": 1.2996,
      "step": 6062
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1170677526841911e-05,
      "loss": 1.2134,
      "step": 6063
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1169196593854127e-05,
      "loss": 1.3122,
      "step": 6064
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1167715660866346e-05,
      "loss": 1.3845,
      "step": 6065
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1166234727878564e-05,
      "loss": 1.152,
      "step": 6066
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1164753794890781e-05,
      "loss": 1.4171,
      "step": 6067
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1163272861903e-05,
      "loss": 1.2984,
      "step": 6068
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1161791928915218e-05,
      "loss": 1.3077,
      "step": 6069
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1160310995927434e-05,
      "loss": 1.1941,
      "step": 6070
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1158830062939653e-05,
      "loss": 1.3285,
      "step": 6071
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.115734912995187e-05,
      "loss": 1.4529,
      "step": 6072
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1155868196964088e-05,
      "loss": 1.6367,
      "step": 6073
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1154387263976306e-05,
      "loss": 1.5881,
      "step": 6074
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1152906330988525e-05,
      "loss": 1.2571,
      "step": 6075
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.115142539800074e-05,
      "loss": 1.3162,
      "step": 6076
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1149944465012958e-05,
      "loss": 1.3132,
      "step": 6077
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1148463532025178e-05,
      "loss": 1.3687,
      "step": 6078
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1146982599037395e-05,
      "loss": 1.0609,
      "step": 6079
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1145501666049613e-05,
      "loss": 1.3246,
      "step": 6080
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1144020733061828e-05,
      "loss": 1.1669,
      "step": 6081
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1142539800074048e-05,
      "loss": 1.0614,
      "step": 6082
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1141058867086265e-05,
      "loss": 1.2223,
      "step": 6083
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1139577934098483e-05,
      "loss": 1.3309,
      "step": 6084
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1138097001110702e-05,
      "loss": 1.2549,
      "step": 6085
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1136616068122918e-05,
      "loss": 1.469,
      "step": 6086
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1135135135135135e-05,
      "loss": 1.1572,
      "step": 6087
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1133654202147354e-05,
      "loss": 1.4635,
      "step": 6088
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1132173269159572e-05,
      "loss": 1.1768,
      "step": 6089
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.113069233617179e-05,
      "loss": 1.4777,
      "step": 6090
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1129211403184005e-05,
      "loss": 1.3352,
      "step": 6091
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1127730470196224e-05,
      "loss": 1.1156,
      "step": 6092
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1126249537208442e-05,
      "loss": 1.5816,
      "step": 6093
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.112476860422066e-05,
      "loss": 1.3958,
      "step": 6094
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1123287671232879e-05,
      "loss": 1.3693,
      "step": 6095
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1121806738245096e-05,
      "loss": 1.2955,
      "step": 6096
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1120325805257312e-05,
      "loss": 1.2034,
      "step": 6097
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.111884487226953e-05,
      "loss": 1.0718,
      "step": 6098
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.1117363939281749e-05,
      "loss": 1.3717,
      "step": 6099
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1115883006293966e-05,
      "loss": 1.4023,
      "step": 6100
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1114402073306184e-05,
      "loss": 1.3195,
      "step": 6101
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1112921140318403e-05,
      "loss": 1.5857,
      "step": 6102
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1111440207330619e-05,
      "loss": 1.6156,
      "step": 6103
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1109959274342836e-05,
      "loss": 1.5599,
      "step": 6104
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1108478341355056e-05,
      "loss": 1.356,
      "step": 6105
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1106997408367273e-05,
      "loss": 1.2091,
      "step": 6106
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1105516475379489e-05,
      "loss": 1.22,
      "step": 6107
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1104035542391706e-05,
      "loss": 1.1222,
      "step": 6108
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1102554609403926e-05,
      "loss": 1.3778,
      "step": 6109
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1101073676416143e-05,
      "loss": 1.2003,
      "step": 6110
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.109959274342836e-05,
      "loss": 1.5575,
      "step": 6111
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.109811181044058e-05,
      "loss": 1.2848,
      "step": 6112
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1096630877452796e-05,
      "loss": 1.2905,
      "step": 6113
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1095149944465013e-05,
      "loss": 1.4627,
      "step": 6114
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.109366901147723e-05,
      "loss": 1.392,
      "step": 6115
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.109218807848945e-05,
      "loss": 1.4077,
      "step": 6116
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1090707145501667e-05,
      "loss": 1.4669,
      "step": 6117
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1089226212513883e-05,
      "loss": 1.3654,
      "step": 6118
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1087745279526103e-05,
      "loss": 1.3612,
      "step": 6119
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.108626434653832e-05,
      "loss": 1.4161,
      "step": 6120
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1084783413550538e-05,
      "loss": 1.2508,
      "step": 6121
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1083302480562757e-05,
      "loss": 1.2352,
      "step": 6122
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1081821547574974e-05,
      "loss": 1.3601,
      "step": 6123
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.108034061458719e-05,
      "loss": 1.4068,
      "step": 6124
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1078859681599408e-05,
      "loss": 1.4197,
      "step": 6125
    },
    {
      "epoch": 1.35,
      "eval_loss": NaN,
      "eval_runtime": 4.3731,
      "eval_samples_per_second": 353.071,
      "eval_steps_per_second": 44.134,
      "step": 6125
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1077378748611627e-05,
      "loss": 1.5638,
      "step": 6126
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1075897815623844e-05,
      "loss": 1.5755,
      "step": 6127
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.107441688263606e-05,
      "loss": 1.2685,
      "step": 6128
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.107293594964828e-05,
      "loss": 1.0702,
      "step": 6129
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1071455016660497e-05,
      "loss": 1.381,
      "step": 6130
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1069974083672714e-05,
      "loss": 1.2041,
      "step": 6131
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1068493150684932e-05,
      "loss": 1.3393,
      "step": 6132
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1067012217697151e-05,
      "loss": 1.4186,
      "step": 6133
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1065531284709367e-05,
      "loss": 1.2916,
      "step": 6134
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1064050351721584e-05,
      "loss": 1.3122,
      "step": 6135
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1062569418733804e-05,
      "loss": 1.3479,
      "step": 6136
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1061088485746021e-05,
      "loss": 1.2372,
      "step": 6137
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1059607552758239e-05,
      "loss": 1.2095,
      "step": 6138
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1058126619770458e-05,
      "loss": 1.3227,
      "step": 6139
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1056645686782674e-05,
      "loss": 1.2742,
      "step": 6140
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1055164753794891e-05,
      "loss": 1.224,
      "step": 6141
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1053683820807109e-05,
      "loss": 1.4823,
      "step": 6142
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1052202887819328e-05,
      "loss": 1.2511,
      "step": 6143
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.1050721954831546e-05,
      "loss": 1.184,
      "step": 6144
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1049241021843761e-05,
      "loss": 1.5384,
      "step": 6145
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.104776008885598e-05,
      "loss": 1.4474,
      "step": 6146
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1046279155868198e-05,
      "loss": 1.4746,
      "step": 6147
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1044798222880416e-05,
      "loss": 1.4341,
      "step": 6148
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1043317289892631e-05,
      "loss": 1.1534,
      "step": 6149
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.104183635690485e-05,
      "loss": 1.3724,
      "step": 6150
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1040355423917068e-05,
      "loss": 1.44,
      "step": 6151
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1038874490929286e-05,
      "loss": 1.059,
      "step": 6152
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1037393557941505e-05,
      "loss": 1.2865,
      "step": 6153
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1035912624953722e-05,
      "loss": 1.4377,
      "step": 6154
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1034431691965938e-05,
      "loss": 1.283,
      "step": 6155
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1032950758978157e-05,
      "loss": 1.2489,
      "step": 6156
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1031469825990375e-05,
      "loss": 1.3415,
      "step": 6157
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1029988893002592e-05,
      "loss": 1.3739,
      "step": 6158
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.102850796001481e-05,
      "loss": 1.3232,
      "step": 6159
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.102702702702703e-05,
      "loss": 1.5736,
      "step": 6160
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1025546094039245e-05,
      "loss": 1.3303,
      "step": 6161
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1024065161051463e-05,
      "loss": 1.3021,
      "step": 6162
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1022584228063682e-05,
      "loss": 1.3175,
      "step": 6163
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.10211032950759e-05,
      "loss": 1.348,
      "step": 6164
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1019622362088117e-05,
      "loss": 1.1961,
      "step": 6165
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1018141429100333e-05,
      "loss": 1.3687,
      "step": 6166
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1016660496112552e-05,
      "loss": 1.3811,
      "step": 6167
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.101517956312477e-05,
      "loss": 1.308,
      "step": 6168
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1013698630136987e-05,
      "loss": 1.2674,
      "step": 6169
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1012217697149206e-05,
      "loss": 1.4509,
      "step": 6170
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1010736764161422e-05,
      "loss": 1.2819,
      "step": 6171
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.100925583117364e-05,
      "loss": 1.2944,
      "step": 6172
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1007774898185859e-05,
      "loss": 1.2919,
      "step": 6173
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1006293965198076e-05,
      "loss": 1.4182,
      "step": 6174
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1004813032210294e-05,
      "loss": 1.2554,
      "step": 6175
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.100333209922251e-05,
      "loss": 1.0373,
      "step": 6176
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1001851166234729e-05,
      "loss": 1.4158,
      "step": 6177
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.1000370233246946e-05,
      "loss": 1.0217,
      "step": 6178
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0998889300259164e-05,
      "loss": 1.4187,
      "step": 6179
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0997408367271383e-05,
      "loss": 1.1504,
      "step": 6180
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.09959274342836e-05,
      "loss": 1.3496,
      "step": 6181
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0994446501295816e-05,
      "loss": 1.2708,
      "step": 6182
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0992965568308034e-05,
      "loss": 1.4306,
      "step": 6183
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0991484635320253e-05,
      "loss": 1.0984,
      "step": 6184
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.099000370233247e-05,
      "loss": 1.2639,
      "step": 6185
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0988522769344688e-05,
      "loss": 1.3439,
      "step": 6186
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0987041836356907e-05,
      "loss": 1.4046,
      "step": 6187
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.0985560903369123e-05,
      "loss": 1.2318,
      "step": 6188
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.098407997038134e-05,
      "loss": 1.2486,
      "step": 6189
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.098259903739356e-05,
      "loss": 1.5054,
      "step": 6190
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0981118104405777e-05,
      "loss": 1.1031,
      "step": 6191
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0979637171417993e-05,
      "loss": 1.4129,
      "step": 6192
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.097815623843021e-05,
      "loss": 1.235,
      "step": 6193
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.097667530544243e-05,
      "loss": 1.2616,
      "step": 6194
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0975194372454647e-05,
      "loss": 1.4418,
      "step": 6195
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0973713439466865e-05,
      "loss": 1.3341,
      "step": 6196
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0972232506479084e-05,
      "loss": 1.427,
      "step": 6197
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.09707515734913e-05,
      "loss": 1.3107,
      "step": 6198
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0969270640503517e-05,
      "loss": 1.3734,
      "step": 6199
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0967789707515737e-05,
      "loss": 1.414,
      "step": 6200
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0966308774527954e-05,
      "loss": 1.3564,
      "step": 6201
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0964827841540172e-05,
      "loss": 1.099,
      "step": 6202
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0963346908552387e-05,
      "loss": 1.1482,
      "step": 6203
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0961865975564607e-05,
      "loss": 1.3061,
      "step": 6204
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0960385042576824e-05,
      "loss": 1.466,
      "step": 6205
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0958904109589042e-05,
      "loss": 1.381,
      "step": 6206
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0957423176601261e-05,
      "loss": 0.9993,
      "step": 6207
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0955942243613478e-05,
      "loss": 1.2624,
      "step": 6208
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0954461310625694e-05,
      "loss": 1.307,
      "step": 6209
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0952980377637912e-05,
      "loss": 1.7441,
      "step": 6210
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0951499444650131e-05,
      "loss": 1.4163,
      "step": 6211
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0950018511662349e-05,
      "loss": 1.1719,
      "step": 6212
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0948537578674564e-05,
      "loss": 1.2442,
      "step": 6213
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0947056645686784e-05,
      "loss": 1.1407,
      "step": 6214
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0945575712699001e-05,
      "loss": 1.2146,
      "step": 6215
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0944094779711219e-05,
      "loss": 1.1858,
      "step": 6216
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0942613846723438e-05,
      "loss": 1.332,
      "step": 6217
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0941132913735655e-05,
      "loss": 1.2368,
      "step": 6218
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0939651980747871e-05,
      "loss": 1.3197,
      "step": 6219
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0938171047760089e-05,
      "loss": 1.1043,
      "step": 6220
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0936690114772308e-05,
      "loss": 1.4288,
      "step": 6221
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0935209181784525e-05,
      "loss": 1.2572,
      "step": 6222
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0933728248796743e-05,
      "loss": 1.5837,
      "step": 6223
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0932247315808962e-05,
      "loss": 1.5191,
      "step": 6224
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0930766382821178e-05,
      "loss": 1.2781,
      "step": 6225
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0929285449833395e-05,
      "loss": 1.2087,
      "step": 6226
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0927804516845613e-05,
      "loss": 1.0535,
      "step": 6227
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0926323583857832e-05,
      "loss": 1.3995,
      "step": 6228
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.092484265087005e-05,
      "loss": 1.1015,
      "step": 6229
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0923361717882266e-05,
      "loss": 1.1388,
      "step": 6230
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0921880784894485e-05,
      "loss": 1.3319,
      "step": 6231
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0920399851906702e-05,
      "loss": 1.1108,
      "step": 6232
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.091891891891892e-05,
      "loss": 1.1477,
      "step": 6233
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0917437985931139e-05,
      "loss": 1.1792,
      "step": 6234
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.0915957052943355e-05,
      "loss": 1.1852,
      "step": 6235
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0914476119955572e-05,
      "loss": 1.2762,
      "step": 6236
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.091299518696779e-05,
      "loss": 1.4383,
      "step": 6237
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0911514253980009e-05,
      "loss": 1.2076,
      "step": 6238
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0910033320992227e-05,
      "loss": 1.3333,
      "step": 6239
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0908552388004442e-05,
      "loss": 1.3688,
      "step": 6240
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0907071455016662e-05,
      "loss": 1.2512,
      "step": 6241
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0905590522028879e-05,
      "loss": 1.5838,
      "step": 6242
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0904109589041097e-05,
      "loss": 1.3022,
      "step": 6243
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0902628656053314e-05,
      "loss": 1.5484,
      "step": 6244
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0901147723065533e-05,
      "loss": 1.5478,
      "step": 6245
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.089966679007775e-05,
      "loss": 1.4236,
      "step": 6246
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0898185857089967e-05,
      "loss": 1.5465,
      "step": 6247
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0896704924102186e-05,
      "loss": 1.1868,
      "step": 6248
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0895223991114403e-05,
      "loss": 1.2175,
      "step": 6249
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0893743058126621e-05,
      "loss": 1.1709,
      "step": 6250
    },
    {
      "epoch": 1.38,
      "eval_loss": NaN,
      "eval_runtime": 4.3759,
      "eval_samples_per_second": 352.844,
      "eval_steps_per_second": 44.105,
      "step": 6250
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.089226212513884e-05,
      "loss": 1.3377,
      "step": 6251
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0890781192151056e-05,
      "loss": 1.3486,
      "step": 6252
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0889300259163274e-05,
      "loss": 1.4118,
      "step": 6253
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0887819326175491e-05,
      "loss": 1.107,
      "step": 6254
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.088633839318771e-05,
      "loss": 1.5876,
      "step": 6255
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0884857460199926e-05,
      "loss": 1.3412,
      "step": 6256
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0883376527212144e-05,
      "loss": 1.299,
      "step": 6257
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0881895594224363e-05,
      "loss": 1.2584,
      "step": 6258
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.088041466123658e-05,
      "loss": 1.1534,
      "step": 6259
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0878933728248798e-05,
      "loss": 1.3329,
      "step": 6260
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0877452795261014e-05,
      "loss": 1.2786,
      "step": 6261
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0875971862273233e-05,
      "loss": 1.3267,
      "step": 6262
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.087449092928545e-05,
      "loss": 1.1836,
      "step": 6263
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0873009996297668e-05,
      "loss": 1.1389,
      "step": 6264
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0871529063309887e-05,
      "loss": 1.4147,
      "step": 6265
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0870048130322105e-05,
      "loss": 1.4838,
      "step": 6266
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.086856719733432e-05,
      "loss": 1.2782,
      "step": 6267
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.086708626434654e-05,
      "loss": 1.4493,
      "step": 6268
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0865605331358757e-05,
      "loss": 1.3743,
      "step": 6269
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0864124398370975e-05,
      "loss": 1.1803,
      "step": 6270
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0862643465383192e-05,
      "loss": 1.391,
      "step": 6271
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0861162532395411e-05,
      "loss": 1.0783,
      "step": 6272
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0859681599407627e-05,
      "loss": 1.4129,
      "step": 6273
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0858200666419845e-05,
      "loss": 1.2716,
      "step": 6274
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0856719733432064e-05,
      "loss": 1.3206,
      "step": 6275
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0855238800444281e-05,
      "loss": 1.0941,
      "step": 6276
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0853757867456497e-05,
      "loss": 1.2249,
      "step": 6277
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0852276934468715e-05,
      "loss": 1.4852,
      "step": 6278
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0850796001480934e-05,
      "loss": 1.375,
      "step": 6279
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.0849315068493152e-05,
      "loss": 1.1274,
      "step": 6280
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0847834135505369e-05,
      "loss": 1.3838,
      "step": 6281
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0846353202517588e-05,
      "loss": 1.3327,
      "step": 6282
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0844872269529804e-05,
      "loss": 1.2022,
      "step": 6283
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0843391336542022e-05,
      "loss": 1.2802,
      "step": 6284
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.084191040355424e-05,
      "loss": 1.2443,
      "step": 6285
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0840429470566458e-05,
      "loss": 1.1852,
      "step": 6286
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0838948537578676e-05,
      "loss": 1.4754,
      "step": 6287
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0837467604590892e-05,
      "loss": 1.4516,
      "step": 6288
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0835986671603111e-05,
      "loss": 1.2449,
      "step": 6289
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0834505738615328e-05,
      "loss": 1.2148,
      "step": 6290
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0833024805627546e-05,
      "loss": 1.1705,
      "step": 6291
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0831543872639765e-05,
      "loss": 1.319,
      "step": 6292
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0830062939651983e-05,
      "loss": 1.3991,
      "step": 6293
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0828582006664198e-05,
      "loss": 1.4355,
      "step": 6294
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0827101073676416e-05,
      "loss": 1.1997,
      "step": 6295
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0825620140688635e-05,
      "loss": 1.1824,
      "step": 6296
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0824139207700853e-05,
      "loss": 1.3004,
      "step": 6297
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0822658274713069e-05,
      "loss": 1.2497,
      "step": 6298
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0821177341725288e-05,
      "loss": 1.2738,
      "step": 6299
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0819696408737505e-05,
      "loss": 1.247,
      "step": 6300
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0818215475749723e-05,
      "loss": 1.3111,
      "step": 6301
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0816734542761942e-05,
      "loss": 1.341,
      "step": 6302
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.081525360977416e-05,
      "loss": 1.4367,
      "step": 6303
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0813772676786375e-05,
      "loss": 1.2379,
      "step": 6304
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0812291743798593e-05,
      "loss": 1.1988,
      "step": 6305
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0810810810810812e-05,
      "loss": 1.1771,
      "step": 6306
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.080932987782303e-05,
      "loss": 1.2889,
      "step": 6307
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0807848944835247e-05,
      "loss": 1.1416,
      "step": 6308
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0806368011847466e-05,
      "loss": 1.3049,
      "step": 6309
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0804887078859682e-05,
      "loss": 1.4004,
      "step": 6310
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.08034061458719e-05,
      "loss": 1.4746,
      "step": 6311
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0801925212884117e-05,
      "loss": 1.3571,
      "step": 6312
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0800444279896336e-05,
      "loss": 1.1997,
      "step": 6313
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0798963346908554e-05,
      "loss": 1.2465,
      "step": 6314
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.079748241392077e-05,
      "loss": 1.2381,
      "step": 6315
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0796001480932989e-05,
      "loss": 1.088,
      "step": 6316
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0794520547945206e-05,
      "loss": 1.3871,
      "step": 6317
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0793039614957424e-05,
      "loss": 1.2103,
      "step": 6318
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0791558681969643e-05,
      "loss": 1.1464,
      "step": 6319
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0790077748981859e-05,
      "loss": 1.5497,
      "step": 6320
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0788596815994077e-05,
      "loss": 1.1527,
      "step": 6321
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0787115883006294e-05,
      "loss": 1.4282,
      "step": 6322
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0785634950018513e-05,
      "loss": 1.1445,
      "step": 6323
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.078415401703073e-05,
      "loss": 1.3554,
      "step": 6324
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0782673084042947e-05,
      "loss": 1.5823,
      "step": 6325
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.0781192151055166e-05,
      "loss": 1.2989,
      "step": 6326
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0779711218067383e-05,
      "loss": 1.507,
      "step": 6327
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.07782302850796e-05,
      "loss": 1.3748,
      "step": 6328
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.077674935209182e-05,
      "loss": 1.2118,
      "step": 6329
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0775268419104038e-05,
      "loss": 1.2187,
      "step": 6330
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0773787486116253e-05,
      "loss": 1.3132,
      "step": 6331
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0772306553128471e-05,
      "loss": 1.2922,
      "step": 6332
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.077082562014069e-05,
      "loss": 1.365,
      "step": 6333
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0769344687152908e-05,
      "loss": 1.3014,
      "step": 6334
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0767863754165125e-05,
      "loss": 1.0648,
      "step": 6335
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0766382821177344e-05,
      "loss": 1.5187,
      "step": 6336
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.076490188818956e-05,
      "loss": 1.442,
      "step": 6337
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0763420955201778e-05,
      "loss": 1.2853,
      "step": 6338
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0761940022213995e-05,
      "loss": 1.4895,
      "step": 6339
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0760459089226214e-05,
      "loss": 1.5608,
      "step": 6340
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.075897815623843e-05,
      "loss": 1.1512,
      "step": 6341
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0757497223250648e-05,
      "loss": 1.1339,
      "step": 6342
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0756016290262867e-05,
      "loss": 1.3484,
      "step": 6343
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0754535357275084e-05,
      "loss": 1.354,
      "step": 6344
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0753054424287302e-05,
      "loss": 1.3403,
      "step": 6345
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0751573491299521e-05,
      "loss": 1.0179,
      "step": 6346
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0750092558311737e-05,
      "loss": 1.0049,
      "step": 6347
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0748611625323955e-05,
      "loss": 1.258,
      "step": 6348
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0747130692336172e-05,
      "loss": 1.3957,
      "step": 6349
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0745649759348391e-05,
      "loss": 1.2484,
      "step": 6350
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0744168826360609e-05,
      "loss": 1.1972,
      "step": 6351
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0742687893372825e-05,
      "loss": 1.3338,
      "step": 6352
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0741206960385044e-05,
      "loss": 1.3321,
      "step": 6353
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0739726027397261e-05,
      "loss": 1.4129,
      "step": 6354
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0738245094409479e-05,
      "loss": 1.1885,
      "step": 6355
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0736764161421696e-05,
      "loss": 1.2435,
      "step": 6356
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0735283228433916e-05,
      "loss": 1.2691,
      "step": 6357
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0733802295446131e-05,
      "loss": 1.338,
      "step": 6358
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0732321362458349e-05,
      "loss": 1.2367,
      "step": 6359
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0730840429470568e-05,
      "loss": 1.5679,
      "step": 6360
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0729359496482786e-05,
      "loss": 1.2062,
      "step": 6361
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0727878563495001e-05,
      "loss": 1.5508,
      "step": 6362
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.072639763050722e-05,
      "loss": 1.2801,
      "step": 6363
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0724916697519438e-05,
      "loss": 1.369,
      "step": 6364
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0723435764531656e-05,
      "loss": 1.4714,
      "step": 6365
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0721954831543873e-05,
      "loss": 1.2855,
      "step": 6366
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0720473898556092e-05,
      "loss": 1.1973,
      "step": 6367
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0718992965568308e-05,
      "loss": 1.0939,
      "step": 6368
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0717512032580526e-05,
      "loss": 1.2609,
      "step": 6369
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0716031099592745e-05,
      "loss": 1.1448,
      "step": 6370
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0714550166604963e-05,
      "loss": 1.2531,
      "step": 6371
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.071306923361718e-05,
      "loss": 1.4239,
      "step": 6372
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0711588300629396e-05,
      "loss": 1.2553,
      "step": 6373
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0710107367641615e-05,
      "loss": 1.5332,
      "step": 6374
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0708626434653833e-05,
      "loss": 1.2477,
      "step": 6375
    },
    {
      "epoch": 1.41,
      "eval_loss": NaN,
      "eval_runtime": 4.3701,
      "eval_samples_per_second": 353.309,
      "eval_steps_per_second": 44.164,
      "step": 6375
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.070714550166605e-05,
      "loss": 1.292,
      "step": 6376
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.070566456867827e-05,
      "loss": 1.6631,
      "step": 6377
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0704183635690487e-05,
      "loss": 1.3745,
      "step": 6378
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0702702702702703e-05,
      "loss": 1.2216,
      "step": 6379
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0701221769714922e-05,
      "loss": 1.0997,
      "step": 6380
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.069974083672714e-05,
      "loss": 1.1884,
      "step": 6381
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0698259903739357e-05,
      "loss": 1.2809,
      "step": 6382
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0696778970751573e-05,
      "loss": 1.2892,
      "step": 6383
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0695298037763792e-05,
      "loss": 1.1918,
      "step": 6384
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.069381710477601e-05,
      "loss": 1.2272,
      "step": 6385
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0692336171788227e-05,
      "loss": 1.5335,
      "step": 6386
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0690855238800446e-05,
      "loss": 1.222,
      "step": 6387
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0689374305812664e-05,
      "loss": 1.3814,
      "step": 6388
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.068789337282488e-05,
      "loss": 1.2809,
      "step": 6389
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0686412439837097e-05,
      "loss": 1.1697,
      "step": 6390
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0684931506849316e-05,
      "loss": 1.5255,
      "step": 6391
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0683450573861534e-05,
      "loss": 0.9803,
      "step": 6392
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0681969640873751e-05,
      "loss": 1.2304,
      "step": 6393
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.068048870788597e-05,
      "loss": 1.4416,
      "step": 6394
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0679007774898186e-05,
      "loss": 1.1894,
      "step": 6395
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0677526841910404e-05,
      "loss": 1.5384,
      "step": 6396
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0676045908922623e-05,
      "loss": 1.0584,
      "step": 6397
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.067456497593484e-05,
      "loss": 1.4772,
      "step": 6398
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0673084042947058e-05,
      "loss": 1.6027,
      "step": 6399
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0671603109959274e-05,
      "loss": 1.2404,
      "step": 6400
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0670122176971493e-05,
      "loss": 1.3178,
      "step": 6401
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.066864124398371e-05,
      "loss": 1.4066,
      "step": 6402
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0667160310995928e-05,
      "loss": 1.2648,
      "step": 6403
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0665679378008147e-05,
      "loss": 1.3675,
      "step": 6404
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0664198445020363e-05,
      "loss": 1.2905,
      "step": 6405
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.066271751203258e-05,
      "loss": 1.1973,
      "step": 6406
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0661236579044798e-05,
      "loss": 1.2409,
      "step": 6407
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0659755646057017e-05,
      "loss": 1.3102,
      "step": 6408
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0658274713069235e-05,
      "loss": 1.3357,
      "step": 6409
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.065679378008145e-05,
      "loss": 1.2984,
      "step": 6410
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.065531284709367e-05,
      "loss": 1.3329,
      "step": 6411
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0653831914105887e-05,
      "loss": 1.0721,
      "step": 6412
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0652350981118105e-05,
      "loss": 0.9412,
      "step": 6413
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0650870048130324e-05,
      "loss": 1.5109,
      "step": 6414
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0649389115142542e-05,
      "loss": 1.5668,
      "step": 6415
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0647908182154758e-05,
      "loss": 1.2535,
      "step": 6416
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.0646427249166975e-05,
      "loss": 1.1961,
      "step": 6417
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0644946316179194e-05,
      "loss": 1.2118,
      "step": 6418
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0643465383191412e-05,
      "loss": 1.7183,
      "step": 6419
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.064198445020363e-05,
      "loss": 1.3586,
      "step": 6420
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0640503517215849e-05,
      "loss": 1.3442,
      "step": 6421
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0639022584228064e-05,
      "loss": 1.2059,
      "step": 6422
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0637541651240282e-05,
      "loss": 1.2813,
      "step": 6423
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.06360607182525e-05,
      "loss": 1.3328,
      "step": 6424
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0634579785264719e-05,
      "loss": 1.4233,
      "step": 6425
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0633098852276934e-05,
      "loss": 1.568,
      "step": 6426
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0631617919289152e-05,
      "loss": 1.1499,
      "step": 6427
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0630136986301371e-05,
      "loss": 1.2793,
      "step": 6428
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0628656053313589e-05,
      "loss": 1.3453,
      "step": 6429
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0627175120325806e-05,
      "loss": 1.2036,
      "step": 6430
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0625694187338025e-05,
      "loss": 1.4632,
      "step": 6431
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0624213254350241e-05,
      "loss": 1.2524,
      "step": 6432
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0622732321362459e-05,
      "loss": 1.4061,
      "step": 6433
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0621251388374676e-05,
      "loss": 1.5274,
      "step": 6434
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0619770455386895e-05,
      "loss": 1.3173,
      "step": 6435
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0618289522399113e-05,
      "loss": 1.0941,
      "step": 6436
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0616808589411329e-05,
      "loss": 1.1126,
      "step": 6437
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0615327656423548e-05,
      "loss": 1.174,
      "step": 6438
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0613846723435766e-05,
      "loss": 1.1104,
      "step": 6439
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0612365790447983e-05,
      "loss": 1.3003,
      "step": 6440
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0610884857460199e-05,
      "loss": 1.3732,
      "step": 6441
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.060940392447242e-05,
      "loss": 1.2835,
      "step": 6442
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0607922991484636e-05,
      "loss": 1.1959,
      "step": 6443
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0606442058496853e-05,
      "loss": 1.4374,
      "step": 6444
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0604961125509072e-05,
      "loss": 1.3981,
      "step": 6445
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.060348019252129e-05,
      "loss": 1.3881,
      "step": 6446
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0601999259533506e-05,
      "loss": 1.6035,
      "step": 6447
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0600518326545725e-05,
      "loss": 1.2475,
      "step": 6448
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0599037393557942e-05,
      "loss": 1.3767,
      "step": 6449
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.059755646057016e-05,
      "loss": 1.294,
      "step": 6450
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0596075527582377e-05,
      "loss": 1.3883,
      "step": 6451
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0594594594594597e-05,
      "loss": 1.1628,
      "step": 6452
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0593113661606812e-05,
      "loss": 1.4043,
      "step": 6453
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.059163272861903e-05,
      "loss": 1.4579,
      "step": 6454
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.059015179563125e-05,
      "loss": 1.5734,
      "step": 6455
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0588670862643467e-05,
      "loss": 1.1616,
      "step": 6456
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0587189929655684e-05,
      "loss": 1.1939,
      "step": 6457
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0585708996667903e-05,
      "loss": 1.2956,
      "step": 6458
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.058422806368012e-05,
      "loss": 1.4369,
      "step": 6459
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0582747130692337e-05,
      "loss": 1.4035,
      "step": 6460
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0581266197704554e-05,
      "loss": 1.2946,
      "step": 6461
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.0579785264716773e-05,
      "loss": 1.1923,
      "step": 6462
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0578304331728991e-05,
      "loss": 1.3465,
      "step": 6463
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0576823398741207e-05,
      "loss": 1.47,
      "step": 6464
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0575342465753426e-05,
      "loss": 1.2567,
      "step": 6465
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0573861532765644e-05,
      "loss": 1.3723,
      "step": 6466
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0572380599777861e-05,
      "loss": 1.2883,
      "step": 6467
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0570899666790077e-05,
      "loss": 1.2984,
      "step": 6468
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0569418733802296e-05,
      "loss": 1.248,
      "step": 6469
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0567937800814514e-05,
      "loss": 1.2803,
      "step": 6470
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0566456867826731e-05,
      "loss": 1.2223,
      "step": 6471
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.056497593483895e-05,
      "loss": 1.1601,
      "step": 6472
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0563495001851168e-05,
      "loss": 1.3225,
      "step": 6473
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0562014068863384e-05,
      "loss": 1.1801,
      "step": 6474
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0560533135875603e-05,
      "loss": 1.3067,
      "step": 6475
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.055905220288782e-05,
      "loss": 1.32,
      "step": 6476
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0557571269900038e-05,
      "loss": 1.4048,
      "step": 6477
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0556090336912255e-05,
      "loss": 1.3633,
      "step": 6478
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0554609403924475e-05,
      "loss": 1.1635,
      "step": 6479
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.055312847093669e-05,
      "loss": 1.6289,
      "step": 6480
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0551647537948908e-05,
      "loss": 1.4244,
      "step": 6481
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0550166604961127e-05,
      "loss": 1.4586,
      "step": 6482
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0548685671973345e-05,
      "loss": 1.141,
      "step": 6483
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.054720473898556e-05,
      "loss": 1.326,
      "step": 6484
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0545723805997778e-05,
      "loss": 1.2469,
      "step": 6485
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0544242873009997e-05,
      "loss": 1.3688,
      "step": 6486
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0542761940022215e-05,
      "loss": 1.4665,
      "step": 6487
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0541281007034432e-05,
      "loss": 1.3493,
      "step": 6488
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0539800074046652e-05,
      "loss": 1.1641,
      "step": 6489
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0538319141058867e-05,
      "loss": 1.5083,
      "step": 6490
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0536838208071085e-05,
      "loss": 1.3795,
      "step": 6491
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0535357275083304e-05,
      "loss": 1.1913,
      "step": 6492
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0533876342095522e-05,
      "loss": 1.1154,
      "step": 6493
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0532395409107739e-05,
      "loss": 1.3905,
      "step": 6494
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0530914476119955e-05,
      "loss": 1.2788,
      "step": 6495
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0529433543132174e-05,
      "loss": 1.4284,
      "step": 6496
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0527952610144392e-05,
      "loss": 1.3664,
      "step": 6497
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.052647167715661e-05,
      "loss": 1.417,
      "step": 6498
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0524990744168828e-05,
      "loss": 1.3903,
      "step": 6499
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0523509811181046e-05,
      "loss": 1.415,
      "step": 6500
    },
    {
      "epoch": 1.43,
      "eval_loss": NaN,
      "eval_runtime": 4.3718,
      "eval_samples_per_second": 353.169,
      "eval_steps_per_second": 44.146,
      "step": 6500
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0522028878193262e-05,
      "loss": 1.3458,
      "step": 6501
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.052054794520548e-05,
      "loss": 1.245,
      "step": 6502
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0519067012217698e-05,
      "loss": 1.2508,
      "step": 6503
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0517586079229916e-05,
      "loss": 1.2012,
      "step": 6504
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0516105146242132e-05,
      "loss": 1.4042,
      "step": 6505
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0514624213254353e-05,
      "loss": 1.0975,
      "step": 6506
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.0513143280266569e-05,
      "loss": 1.2701,
      "step": 6507
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0511662347278786e-05,
      "loss": 1.2347,
      "step": 6508
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0510181414291005e-05,
      "loss": 1.5792,
      "step": 6509
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0508700481303223e-05,
      "loss": 1.2265,
      "step": 6510
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0507219548315439e-05,
      "loss": 1.3542,
      "step": 6511
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0505738615327656e-05,
      "loss": 1.2952,
      "step": 6512
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0504257682339875e-05,
      "loss": 1.1364,
      "step": 6513
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0502776749352093e-05,
      "loss": 1.6139,
      "step": 6514
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.050129581636431e-05,
      "loss": 1.2476,
      "step": 6515
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.049981488337653e-05,
      "loss": 1.1852,
      "step": 6516
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0498333950388745e-05,
      "loss": 1.2955,
      "step": 6517
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0496853017400963e-05,
      "loss": 1.2873,
      "step": 6518
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.049537208441318e-05,
      "loss": 1.2491,
      "step": 6519
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.04938911514254e-05,
      "loss": 1.3462,
      "step": 6520
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0492410218437617e-05,
      "loss": 1.4465,
      "step": 6521
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0490929285449833e-05,
      "loss": 1.208,
      "step": 6522
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0489448352462052e-05,
      "loss": 1.3685,
      "step": 6523
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.048796741947427e-05,
      "loss": 1.3233,
      "step": 6524
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0486486486486487e-05,
      "loss": 1.1978,
      "step": 6525
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0485005553498706e-05,
      "loss": 1.3623,
      "step": 6526
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0483524620510924e-05,
      "loss": 1.4088,
      "step": 6527
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.048204368752314e-05,
      "loss": 1.3071,
      "step": 6528
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0480562754535357e-05,
      "loss": 1.7442,
      "step": 6529
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0479081821547577e-05,
      "loss": 1.2447,
      "step": 6530
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0477600888559794e-05,
      "loss": 1.2114,
      "step": 6531
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.047611995557201e-05,
      "loss": 1.2612,
      "step": 6532
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0474639022584229e-05,
      "loss": 1.3463,
      "step": 6533
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0473158089596447e-05,
      "loss": 1.2367,
      "step": 6534
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0471677156608664e-05,
      "loss": 1.2532,
      "step": 6535
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0470196223620882e-05,
      "loss": 1.1643,
      "step": 6536
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.04687152906331e-05,
      "loss": 1.4957,
      "step": 6537
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0467234357645317e-05,
      "loss": 1.2415,
      "step": 6538
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0465753424657534e-05,
      "loss": 1.3237,
      "step": 6539
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0464272491669753e-05,
      "loss": 1.2187,
      "step": 6540
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0462791558681971e-05,
      "loss": 1.2372,
      "step": 6541
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0461310625694188e-05,
      "loss": 1.403,
      "step": 6542
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0459829692706408e-05,
      "loss": 1.293,
      "step": 6543
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0458348759718623e-05,
      "loss": 1.39,
      "step": 6544
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0456867826730841e-05,
      "loss": 1.1689,
      "step": 6545
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0455386893743058e-05,
      "loss": 1.1186,
      "step": 6546
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0453905960755278e-05,
      "loss": 1.2539,
      "step": 6547
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0452425027767494e-05,
      "loss": 1.3018,
      "step": 6548
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0450944094779711e-05,
      "loss": 1.4254,
      "step": 6549
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.044946316179193e-05,
      "loss": 1.3026,
      "step": 6550
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0447982228804148e-05,
      "loss": 1.3913,
      "step": 6551
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0446501295816365e-05,
      "loss": 1.4786,
      "step": 6552
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0445020362828581e-05,
      "loss": 1.4167,
      "step": 6553
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.04435394298408e-05,
      "loss": 1.1175,
      "step": 6554
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0442058496853018e-05,
      "loss": 1.2716,
      "step": 6555
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0440577563865235e-05,
      "loss": 1.1435,
      "step": 6556
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0439096630877455e-05,
      "loss": 1.2006,
      "step": 6557
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0437615697889672e-05,
      "loss": 1.1008,
      "step": 6558
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0436134764901888e-05,
      "loss": 1.431,
      "step": 6559
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0434653831914107e-05,
      "loss": 1.3028,
      "step": 6560
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0433172898926325e-05,
      "loss": 1.2385,
      "step": 6561
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0431691965938542e-05,
      "loss": 1.2994,
      "step": 6562
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.043021103295076e-05,
      "loss": 1.258,
      "step": 6563
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0428730099962979e-05,
      "loss": 1.2652,
      "step": 6564
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0427249166975195e-05,
      "loss": 1.2721,
      "step": 6565
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0425768233987412e-05,
      "loss": 1.5365,
      "step": 6566
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0424287300999631e-05,
      "loss": 1.0851,
      "step": 6567
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0422806368011849e-05,
      "loss": 1.3226,
      "step": 6568
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0421325435024065e-05,
      "loss": 1.3688,
      "step": 6569
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0419844502036282e-05,
      "loss": 1.1525,
      "step": 6570
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0418363569048501e-05,
      "loss": 1.2369,
      "step": 6571
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0416882636060719e-05,
      "loss": 1.4338,
      "step": 6572
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0415401703072937e-05,
      "loss": 1.275,
      "step": 6573
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0413920770085156e-05,
      "loss": 1.5376,
      "step": 6574
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0412439837097372e-05,
      "loss": 1.2292,
      "step": 6575
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0410958904109589e-05,
      "loss": 1.4227,
      "step": 6576
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0409477971121808e-05,
      "loss": 1.4099,
      "step": 6577
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0407997038134026e-05,
      "loss": 1.156,
      "step": 6578
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0406516105146243e-05,
      "loss": 1.4192,
      "step": 6579
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0405035172158459e-05,
      "loss": 1.0805,
      "step": 6580
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0403554239170678e-05,
      "loss": 1.5177,
      "step": 6581
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0402073306182896e-05,
      "loss": 1.3331,
      "step": 6582
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0400592373195113e-05,
      "loss": 1.4409,
      "step": 6583
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0399111440207333e-05,
      "loss": 1.4727,
      "step": 6584
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.039763050721955e-05,
      "loss": 1.342,
      "step": 6585
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0396149574231766e-05,
      "loss": 1.3648,
      "step": 6586
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0394668641243985e-05,
      "loss": 1.2866,
      "step": 6587
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0393187708256203e-05,
      "loss": 1.2176,
      "step": 6588
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.039170677526842e-05,
      "loss": 1.3985,
      "step": 6589
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0390225842280636e-05,
      "loss": 1.3576,
      "step": 6590
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0388744909292857e-05,
      "loss": 1.2632,
      "step": 6591
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0387263976305073e-05,
      "loss": 1.2658,
      "step": 6592
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.038578304331729e-05,
      "loss": 1.1618,
      "step": 6593
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.038430211032951e-05,
      "loss": 1.4142,
      "step": 6594
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0382821177341727e-05,
      "loss": 1.4718,
      "step": 6595
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.0381340244353943e-05,
      "loss": 1.2779,
      "step": 6596
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.037985931136616e-05,
      "loss": 1.1845,
      "step": 6597
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.037837837837838e-05,
      "loss": 1.3389,
      "step": 6598
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0376897445390597e-05,
      "loss": 1.3347,
      "step": 6599
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0375416512402815e-05,
      "loss": 1.4348,
      "step": 6600
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0373935579415034e-05,
      "loss": 1.1793,
      "step": 6601
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.037245464642725e-05,
      "loss": 1.1471,
      "step": 6602
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0370973713439467e-05,
      "loss": 1.2723,
      "step": 6603
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0369492780451686e-05,
      "loss": 1.1702,
      "step": 6604
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0368011847463904e-05,
      "loss": 1.2713,
      "step": 6605
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0366530914476121e-05,
      "loss": 1.1311,
      "step": 6606
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0365049981488337e-05,
      "loss": 1.2775,
      "step": 6607
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0363569048500556e-05,
      "loss": 1.2515,
      "step": 6608
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0362088115512774e-05,
      "loss": 1.308,
      "step": 6609
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0360607182524991e-05,
      "loss": 1.4177,
      "step": 6610
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.035912624953721e-05,
      "loss": 1.2283,
      "step": 6611
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0357645316549426e-05,
      "loss": 1.0452,
      "step": 6612
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0356164383561644e-05,
      "loss": 1.2585,
      "step": 6613
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0354683450573861e-05,
      "loss": 1.2694,
      "step": 6614
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.035320251758608e-05,
      "loss": 1.4824,
      "step": 6615
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0351721584598298e-05,
      "loss": 1.2977,
      "step": 6616
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0350240651610514e-05,
      "loss": 1.2104,
      "step": 6617
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0348759718622733e-05,
      "loss": 1.0001,
      "step": 6618
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.034727878563495e-05,
      "loss": 1.2747,
      "step": 6619
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0345797852647168e-05,
      "loss": 1.5092,
      "step": 6620
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0344316919659387e-05,
      "loss": 1.0971,
      "step": 6621
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0342835986671605e-05,
      "loss": 1.2785,
      "step": 6622
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.034135505368382e-05,
      "loss": 1.1447,
      "step": 6623
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0339874120696038e-05,
      "loss": 1.4395,
      "step": 6624
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0338393187708258e-05,
      "loss": 1.3721,
      "step": 6625
    },
    {
      "epoch": 1.46,
      "eval_loss": NaN,
      "eval_runtime": 4.3712,
      "eval_samples_per_second": 353.224,
      "eval_steps_per_second": 44.153,
      "step": 6625
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0336912254720475e-05,
      "loss": 1.2214,
      "step": 6626
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0335431321732693e-05,
      "loss": 1.4509,
      "step": 6627
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0333950388744912e-05,
      "loss": 1.1788,
      "step": 6628
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0332469455757128e-05,
      "loss": 1.5035,
      "step": 6629
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0330988522769345e-05,
      "loss": 1.3076,
      "step": 6630
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0329507589781563e-05,
      "loss": 1.3717,
      "step": 6631
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0328026656793782e-05,
      "loss": 1.3334,
      "step": 6632
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0326545723805998e-05,
      "loss": 1.1849,
      "step": 6633
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0325064790818215e-05,
      "loss": 1.3502,
      "step": 6634
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0323583857830434e-05,
      "loss": 1.279,
      "step": 6635
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0322102924842652e-05,
      "loss": 1.0451,
      "step": 6636
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.032062199185487e-05,
      "loss": 1.4524,
      "step": 6637
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0319141058867089e-05,
      "loss": 1.2872,
      "step": 6638
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0317660125879304e-05,
      "loss": 1.5266,
      "step": 6639
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0316179192891522e-05,
      "loss": 1.3337,
      "step": 6640
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.031469825990374e-05,
      "loss": 1.1547,
      "step": 6641
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0313217326915959e-05,
      "loss": 1.0545,
      "step": 6642
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.0311736393928176e-05,
      "loss": 1.3643,
      "step": 6643
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0310255460940392e-05,
      "loss": 1.3556,
      "step": 6644
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0308774527952611e-05,
      "loss": 1.3474,
      "step": 6645
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0307293594964829e-05,
      "loss": 1.321,
      "step": 6646
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0305812661977046e-05,
      "loss": 1.0757,
      "step": 6647
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0304331728989264e-05,
      "loss": 1.2719,
      "step": 6648
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0302850796001483e-05,
      "loss": 1.3261,
      "step": 6649
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0301369863013699e-05,
      "loss": 1.4532,
      "step": 6650
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0299888930025916e-05,
      "loss": 1.1768,
      "step": 6651
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0298407997038136e-05,
      "loss": 1.353,
      "step": 6652
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0296927064050353e-05,
      "loss": 1.2896,
      "step": 6653
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0295446131062569e-05,
      "loss": 1.3066,
      "step": 6654
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.029396519807479e-05,
      "loss": 1.1801,
      "step": 6655
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0292484265087006e-05,
      "loss": 1.2124,
      "step": 6656
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0291003332099223e-05,
      "loss": 1.4288,
      "step": 6657
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.028952239911144e-05,
      "loss": 1.2557,
      "step": 6658
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.028804146612366e-05,
      "loss": 1.0622,
      "step": 6659
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0286560533135876e-05,
      "loss": 1.2048,
      "step": 6660
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0285079600148093e-05,
      "loss": 1.3648,
      "step": 6661
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0283598667160312e-05,
      "loss": 1.3905,
      "step": 6662
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.028211773417253e-05,
      "loss": 1.2877,
      "step": 6663
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0280636801184747e-05,
      "loss": 1.2718,
      "step": 6664
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0279155868196963e-05,
      "loss": 1.1321,
      "step": 6665
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0277674935209183e-05,
      "loss": 1.0877,
      "step": 6666
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.02761940022214e-05,
      "loss": 1.5029,
      "step": 6667
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0274713069233618e-05,
      "loss": 1.4272,
      "step": 6668
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0273232136245837e-05,
      "loss": 1.1874,
      "step": 6669
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0271751203258054e-05,
      "loss": 1.2741,
      "step": 6670
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.027027027027027e-05,
      "loss": 1.3892,
      "step": 6671
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.026878933728249e-05,
      "loss": 1.3095,
      "step": 6672
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0267308404294707e-05,
      "loss": 1.2722,
      "step": 6673
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0265827471306924e-05,
      "loss": 1.103,
      "step": 6674
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.026434653831914e-05,
      "loss": 1.0825,
      "step": 6675
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.026286560533136e-05,
      "loss": 1.3438,
      "step": 6676
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0261384672343577e-05,
      "loss": 1.2191,
      "step": 6677
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0259903739355794e-05,
      "loss": 1.1633,
      "step": 6678
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0258422806368014e-05,
      "loss": 1.3943,
      "step": 6679
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0256941873380231e-05,
      "loss": 1.2579,
      "step": 6680
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0255460940392447e-05,
      "loss": 1.3896,
      "step": 6681
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0253980007404664e-05,
      "loss": 1.4798,
      "step": 6682
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0252499074416884e-05,
      "loss": 1.1175,
      "step": 6683
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0251018141429101e-05,
      "loss": 1.1662,
      "step": 6684
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0249537208441319e-05,
      "loss": 1.0148,
      "step": 6685
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0248056275453538e-05,
      "loss": 1.2595,
      "step": 6686
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0246575342465754e-05,
      "loss": 1.0901,
      "step": 6687
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0245094409477971e-05,
      "loss": 1.1385,
      "step": 6688
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.024361347649019e-05,
      "loss": 1.2709,
      "step": 6689
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0242132543502408e-05,
      "loss": 1.3235,
      "step": 6690
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0240651610514626e-05,
      "loss": 1.2747,
      "step": 6691
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0239170677526841e-05,
      "loss": 1.2905,
      "step": 6692
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.023768974453906e-05,
      "loss": 1.6482,
      "step": 6693
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0236208811551278e-05,
      "loss": 1.3507,
      "step": 6694
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0234727878563496e-05,
      "loss": 1.2213,
      "step": 6695
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0233246945575715e-05,
      "loss": 1.2534,
      "step": 6696
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.023176601258793e-05,
      "loss": 1.3282,
      "step": 6697
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0230285079600148e-05,
      "loss": 1.3013,
      "step": 6698
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0228804146612366e-05,
      "loss": 1.156,
      "step": 6699
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0227323213624585e-05,
      "loss": 1.3117,
      "step": 6700
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0225842280636802e-05,
      "loss": 1.3543,
      "step": 6701
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0224361347649018e-05,
      "loss": 1.5459,
      "step": 6702
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0222880414661237e-05,
      "loss": 1.3863,
      "step": 6703
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0221399481673455e-05,
      "loss": 1.2479,
      "step": 6704
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0219918548685672e-05,
      "loss": 1.0909,
      "step": 6705
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0218437615697892e-05,
      "loss": 1.4262,
      "step": 6706
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.021695668271011e-05,
      "loss": 1.3259,
      "step": 6707
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0215475749722325e-05,
      "loss": 1.4679,
      "step": 6708
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0213994816734543e-05,
      "loss": 1.4046,
      "step": 6709
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0212513883746762e-05,
      "loss": 1.2271,
      "step": 6710
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.021103295075898e-05,
      "loss": 1.2791,
      "step": 6711
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0209552017771197e-05,
      "loss": 1.6373,
      "step": 6712
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0208071084783416e-05,
      "loss": 1.2056,
      "step": 6713
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0206590151795632e-05,
      "loss": 1.3282,
      "step": 6714
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.020510921880785e-05,
      "loss": 1.1856,
      "step": 6715
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0203628285820067e-05,
      "loss": 1.3581,
      "step": 6716
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0202147352832286e-05,
      "loss": 1.246,
      "step": 6717
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0200666419844502e-05,
      "loss": 1.3807,
      "step": 6718
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.019918548685672e-05,
      "loss": 1.3524,
      "step": 6719
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0197704553868939e-05,
      "loss": 1.2048,
      "step": 6720
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0196223620881156e-05,
      "loss": 1.191,
      "step": 6721
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0194742687893374e-05,
      "loss": 1.2692,
      "step": 6722
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0193261754905593e-05,
      "loss": 1.3178,
      "step": 6723
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0191780821917809e-05,
      "loss": 1.3464,
      "step": 6724
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0190299888930026e-05,
      "loss": 1.2168,
      "step": 6725
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0188818955942244e-05,
      "loss": 1.4873,
      "step": 6726
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0187338022954463e-05,
      "loss": 1.4968,
      "step": 6727
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.018585708996668e-05,
      "loss": 1.2071,
      "step": 6728
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0184376156978896e-05,
      "loss": 1.3451,
      "step": 6729
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0182895223991115e-05,
      "loss": 1.4235,
      "step": 6730
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0181414291003333e-05,
      "loss": 1.1531,
      "step": 6731
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.017993335801555e-05,
      "loss": 1.3704,
      "step": 6732
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.017845242502777e-05,
      "loss": 1.2944,
      "step": 6733
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0176971492039987e-05,
      "loss": 1.3099,
      "step": 6734
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0175490559052203e-05,
      "loss": 1.1942,
      "step": 6735
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.017400962606442e-05,
      "loss": 1.1794,
      "step": 6736
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.017252869307664e-05,
      "loss": 1.4109,
      "step": 6737
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0171047760088857e-05,
      "loss": 1.2541,
      "step": 6738
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0169566827101073e-05,
      "loss": 1.3102,
      "step": 6739
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0168085894113292e-05,
      "loss": 1.3041,
      "step": 6740
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.016660496112551e-05,
      "loss": 1.1986,
      "step": 6741
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0165124028137727e-05,
      "loss": 1.4781,
      "step": 6742
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0163643095149945e-05,
      "loss": 1.3743,
      "step": 6743
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0162162162162164e-05,
      "loss": 1.3105,
      "step": 6744
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.016068122917438e-05,
      "loss": 1.3284,
      "step": 6745
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0159200296186597e-05,
      "loss": 1.3513,
      "step": 6746
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0157719363198817e-05,
      "loss": 1.2832,
      "step": 6747
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0156238430211034e-05,
      "loss": 1.283,
      "step": 6748
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0154757497223252e-05,
      "loss": 1.1857,
      "step": 6749
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0153276564235471e-05,
      "loss": 1.4394,
      "step": 6750
    },
    {
      "epoch": 1.49,
      "eval_loss": NaN,
      "eval_runtime": 4.375,
      "eval_samples_per_second": 352.915,
      "eval_steps_per_second": 44.114,
      "step": 6750
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0151795631247687e-05,
      "loss": 1.5022,
      "step": 6751
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0150314698259904e-05,
      "loss": 1.5412,
      "step": 6752
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0148833765272122e-05,
      "loss": 1.5144,
      "step": 6753
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0147352832284341e-05,
      "loss": 1.1944,
      "step": 6754
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0145871899296558e-05,
      "loss": 1.3284,
      "step": 6755
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0144390966308774e-05,
      "loss": 0.9951,
      "step": 6756
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0142910033320993e-05,
      "loss": 1.2045,
      "step": 6757
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0141429100333211e-05,
      "loss": 1.1612,
      "step": 6758
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0139948167345429e-05,
      "loss": 1.5133,
      "step": 6759
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0138467234357644e-05,
      "loss": 1.4401,
      "step": 6760
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0136986301369864e-05,
      "loss": 1.4853,
      "step": 6761
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0135505368382081e-05,
      "loss": 1.5749,
      "step": 6762
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0134024435394299e-05,
      "loss": 1.3233,
      "step": 6763
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0132543502406518e-05,
      "loss": 1.1913,
      "step": 6764
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0131062569418735e-05,
      "loss": 1.2608,
      "step": 6765
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0129581636430951e-05,
      "loss": 1.0556,
      "step": 6766
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.012810070344317e-05,
      "loss": 1.2377,
      "step": 6767
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0126619770455388e-05,
      "loss": 1.2708,
      "step": 6768
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0125138837467605e-05,
      "loss": 1.6683,
      "step": 6769
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0123657904479823e-05,
      "loss": 1.3659,
      "step": 6770
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0122176971492042e-05,
      "loss": 1.7584,
      "step": 6771
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0120696038504258e-05,
      "loss": 1.4212,
      "step": 6772
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0119215105516475e-05,
      "loss": 1.2945,
      "step": 6773
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0117734172528695e-05,
      "loss": 1.3294,
      "step": 6774
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0116253239540912e-05,
      "loss": 1.3936,
      "step": 6775
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.011477230655313e-05,
      "loss": 1.398,
      "step": 6776
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0113291373565346e-05,
      "loss": 1.2634,
      "step": 6777
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0111810440577565e-05,
      "loss": 1.3048,
      "step": 6778
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.0110329507589782e-05,
      "loss": 1.2326,
      "step": 6779
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0108848574602e-05,
      "loss": 1.4357,
      "step": 6780
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0107367641614219e-05,
      "loss": 1.4731,
      "step": 6781
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0105886708626435e-05,
      "loss": 1.3946,
      "step": 6782
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0104405775638652e-05,
      "loss": 0.9775,
      "step": 6783
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0102924842650872e-05,
      "loss": 1.3273,
      "step": 6784
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0101443909663089e-05,
      "loss": 1.3467,
      "step": 6785
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0099962976675307e-05,
      "loss": 1.225,
      "step": 6786
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0098482043687522e-05,
      "loss": 1.4709,
      "step": 6787
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0097001110699742e-05,
      "loss": 1.4801,
      "step": 6788
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0095520177711959e-05,
      "loss": 1.1891,
      "step": 6789
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0094039244724177e-05,
      "loss": 1.4608,
      "step": 6790
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0092558311736396e-05,
      "loss": 1.4185,
      "step": 6791
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0091077378748613e-05,
      "loss": 1.2842,
      "step": 6792
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.008959644576083e-05,
      "loss": 1.4062,
      "step": 6793
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0088115512773047e-05,
      "loss": 1.2843,
      "step": 6794
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0086634579785266e-05,
      "loss": 1.3855,
      "step": 6795
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0085153646797483e-05,
      "loss": 1.4216,
      "step": 6796
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0083672713809701e-05,
      "loss": 1.2581,
      "step": 6797
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.008219178082192e-05,
      "loss": 1.0744,
      "step": 6798
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0080710847834136e-05,
      "loss": 1.1772,
      "step": 6799
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0079229914846353e-05,
      "loss": 1.3018,
      "step": 6800
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0077748981858573e-05,
      "loss": 1.387,
      "step": 6801
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.007626804887079e-05,
      "loss": 1.5495,
      "step": 6802
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0074787115883006e-05,
      "loss": 1.4959,
      "step": 6803
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0073306182895224e-05,
      "loss": 1.3058,
      "step": 6804
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0071825249907443e-05,
      "loss": 1.44,
      "step": 6805
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.007034431691966e-05,
      "loss": 1.5069,
      "step": 6806
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0068863383931878e-05,
      "loss": 1.4918,
      "step": 6807
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0067382450944097e-05,
      "loss": 1.2893,
      "step": 6808
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0065901517956313e-05,
      "loss": 1.4709,
      "step": 6809
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.006442058496853e-05,
      "loss": 1.3119,
      "step": 6810
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0062939651980748e-05,
      "loss": 1.4809,
      "step": 6811
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0061458718992967e-05,
      "loss": 1.3562,
      "step": 6812
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0059977786005185e-05,
      "loss": 1.1361,
      "step": 6813
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.00584968530174e-05,
      "loss": 1.3534,
      "step": 6814
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.005701592002962e-05,
      "loss": 1.383,
      "step": 6815
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0055534987041837e-05,
      "loss": 1.208,
      "step": 6816
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0054054054054055e-05,
      "loss": 1.1195,
      "step": 6817
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0052573121066274e-05,
      "loss": 1.5469,
      "step": 6818
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0051092188078491e-05,
      "loss": 1.4785,
      "step": 6819
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0049611255090707e-05,
      "loss": 1.468,
      "step": 6820
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0048130322102925e-05,
      "loss": 1.0491,
      "step": 6821
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0046649389115144e-05,
      "loss": 1.3664,
      "step": 6822
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0045168456127361e-05,
      "loss": 1.2533,
      "step": 6823
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0043687523139577e-05,
      "loss": 1.2371,
      "step": 6824
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0042206590151797e-05,
      "loss": 1.4935,
      "step": 6825
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0040725657164014e-05,
      "loss": 1.3307,
      "step": 6826
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0039244724176232e-05,
      "loss": 1.4426,
      "step": 6827
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0037763791188449e-05,
      "loss": 1.1568,
      "step": 6828
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0036282858200668e-05,
      "loss": 1.2071,
      "step": 6829
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0034801925212884e-05,
      "loss": 1.3537,
      "step": 6830
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0033320992225102e-05,
      "loss": 1.4202,
      "step": 6831
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.003184005923732e-05,
      "loss": 1.2418,
      "step": 6832
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0030359126249538e-05,
      "loss": 1.2772,
      "step": 6833
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0028878193261756e-05,
      "loss": 1.2109,
      "step": 6834
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0027397260273975e-05,
      "loss": 1.1609,
      "step": 6835
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0025916327286191e-05,
      "loss": 1.1399,
      "step": 6836
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0024435394298408e-05,
      "loss": 1.283,
      "step": 6837
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0022954461310626e-05,
      "loss": 1.3908,
      "step": 6838
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0021473528322845e-05,
      "loss": 1.2086,
      "step": 6839
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0019992595335063e-05,
      "loss": 1.3646,
      "step": 6840
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0018511662347278e-05,
      "loss": 1.2641,
      "step": 6841
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0017030729359498e-05,
      "loss": 1.2982,
      "step": 6842
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0015549796371715e-05,
      "loss": 1.5196,
      "step": 6843
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0014068863383933e-05,
      "loss": 1.2468,
      "step": 6844
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0012587930396149e-05,
      "loss": 1.0428,
      "step": 6845
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0011106997408368e-05,
      "loss": 1.0261,
      "step": 6846
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0009626064420585e-05,
      "loss": 1.0945,
      "step": 6847
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0008145131432803e-05,
      "loss": 1.3564,
      "step": 6848
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0006664198445022e-05,
      "loss": 1.3383,
      "step": 6849
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.000518326545724e-05,
      "loss": 1.264,
      "step": 6850
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0003702332469455e-05,
      "loss": 1.2044,
      "step": 6851
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0002221399481675e-05,
      "loss": 1.4108,
      "step": 6852
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0000740466493892e-05,
      "loss": 1.444,
      "step": 6853
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.99925953350611e-06,
      "loss": 1.1618,
      "step": 6854
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.997778600518327e-06,
      "loss": 1.2833,
      "step": 6855
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.996297667530545e-06,
      "loss": 1.3087,
      "step": 6856
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.994816734542762e-06,
      "loss": 1.2221,
      "step": 6857
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.993335801554981e-06,
      "loss": 1.4434,
      "step": 6858
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.991854868567197e-06,
      "loss": 1.3398,
      "step": 6859
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.990373935579416e-06,
      "loss": 1.2425,
      "step": 6860
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.988893002591634e-06,
      "loss": 1.1292,
      "step": 6861
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.987412069603851e-06,
      "loss": 1.2177,
      "step": 6862
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.985931136616069e-06,
      "loss": 1.4024,
      "step": 6863
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.984450203628286e-06,
      "loss": 1.1961,
      "step": 6864
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.982969270640504e-06,
      "loss": 1.2224,
      "step": 6865
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.981488337652721e-06,
      "loss": 1.1871,
      "step": 6866
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.980007404664939e-06,
      "loss": 1.352,
      "step": 6867
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.978526471677158e-06,
      "loss": 1.1703,
      "step": 6868
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.977045538689376e-06,
      "loss": 1.3086,
      "step": 6869
    },
    {
      "epoch": 1.51,
      "learning_rate": 9.975564605701593e-06,
      "loss": 1.4887,
      "step": 6870
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.97408367271381e-06,
      "loss": 1.4161,
      "step": 6871
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.972602739726028e-06,
      "loss": 1.4228,
      "step": 6872
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.971121806738246e-06,
      "loss": 1.3616,
      "step": 6873
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.969640873750463e-06,
      "loss": 1.4225,
      "step": 6874
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.968159940762683e-06,
      "loss": 1.3715,
      "step": 6875
    },
    {
      "epoch": 1.52,
      "eval_loss": NaN,
      "eval_runtime": 4.3732,
      "eval_samples_per_second": 353.056,
      "eval_steps_per_second": 44.132,
      "step": 6875
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.966679007774898e-06,
      "loss": 1.2849,
      "step": 6876
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.965198074787118e-06,
      "loss": 1.2984,
      "step": 6877
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.963717141799333e-06,
      "loss": 1.235,
      "step": 6878
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.962236208811553e-06,
      "loss": 1.281,
      "step": 6879
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.96075527582377e-06,
      "loss": 1.1367,
      "step": 6880
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.959274342835988e-06,
      "loss": 1.3098,
      "step": 6881
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.957793409848205e-06,
      "loss": 1.3275,
      "step": 6882
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.956312476860423e-06,
      "loss": 1.2854,
      "step": 6883
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.95483154387264e-06,
      "loss": 1.2745,
      "step": 6884
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.95335061088486e-06,
      "loss": 1.0754,
      "step": 6885
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.951869677897075e-06,
      "loss": 1.4854,
      "step": 6886
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.950388744909294e-06,
      "loss": 1.3177,
      "step": 6887
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.94890781192151e-06,
      "loss": 1.3028,
      "step": 6888
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.94742687893373e-06,
      "loss": 1.3637,
      "step": 6889
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.945945945945947e-06,
      "loss": 1.4307,
      "step": 6890
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.944465012958164e-06,
      "loss": 1.1586,
      "step": 6891
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.942984079970382e-06,
      "loss": 1.3221,
      "step": 6892
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.9415031469826e-06,
      "loss": 1.5139,
      "step": 6893
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.940022213994817e-06,
      "loss": 1.3582,
      "step": 6894
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.938541281007035e-06,
      "loss": 1.3432,
      "step": 6895
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.937060348019252e-06,
      "loss": 1.1712,
      "step": 6896
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.935579415031471e-06,
      "loss": 1.5361,
      "step": 6897
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.934098482043689e-06,
      "loss": 1.1422,
      "step": 6898
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.932617549055906e-06,
      "loss": 1.2774,
      "step": 6899
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.931136616068124e-06,
      "loss": 1.5669,
      "step": 6900
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.929655683080341e-06,
      "loss": 1.2335,
      "step": 6901
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.928174750092559e-06,
      "loss": 1.5088,
      "step": 6902
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.926693817104776e-06,
      "loss": 1.3376,
      "step": 6903
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.925212884116996e-06,
      "loss": 1.1982,
      "step": 6904
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.923731951129211e-06,
      "loss": 1.3864,
      "step": 6905
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.92225101814143e-06,
      "loss": 1.3799,
      "step": 6906
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.920770085153648e-06,
      "loss": 1.2466,
      "step": 6907
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.919289152165866e-06,
      "loss": 1.5011,
      "step": 6908
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.917808219178083e-06,
      "loss": 1.2299,
      "step": 6909
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.9163272861903e-06,
      "loss": 1.5404,
      "step": 6910
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.914846353202518e-06,
      "loss": 1.144,
      "step": 6911
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.913365420214736e-06,
      "loss": 1.5055,
      "step": 6912
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.911884487226953e-06,
      "loss": 1.343,
      "step": 6913
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.910403554239172e-06,
      "loss": 1.2416,
      "step": 6914
    },
    {
      "epoch": 1.52,
      "learning_rate": 9.908922621251388e-06,
      "loss": 1.3217,
      "step": 6915
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.907441688263607e-06,
      "loss": 1.2626,
      "step": 6916
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.905960755275823e-06,
      "loss": 1.1293,
      "step": 6917
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.904479822288043e-06,
      "loss": 1.2319,
      "step": 6918
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.90299888930026e-06,
      "loss": 1.1905,
      "step": 6919
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.901517956312478e-06,
      "loss": 1.1461,
      "step": 6920
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.900037023324695e-06,
      "loss": 1.3451,
      "step": 6921
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.898556090336913e-06,
      "loss": 1.1203,
      "step": 6922
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.89707515734913e-06,
      "loss": 1.5194,
      "step": 6923
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.89559422436135e-06,
      "loss": 1.3443,
      "step": 6924
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.894113291373567e-06,
      "loss": 1.1645,
      "step": 6925
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.892632358385784e-06,
      "loss": 1.1986,
      "step": 6926
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.891151425398002e-06,
      "loss": 1.4383,
      "step": 6927
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.88967049241022e-06,
      "loss": 1.2646,
      "step": 6928
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.888189559422437e-06,
      "loss": 1.2328,
      "step": 6929
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.886708626434654e-06,
      "loss": 1.1668,
      "step": 6930
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.885227693446872e-06,
      "loss": 1.236,
      "step": 6931
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.88374676045909e-06,
      "loss": 1.4996,
      "step": 6932
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.882265827471309e-06,
      "loss": 1.3273,
      "step": 6933
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.880784894483524e-06,
      "loss": 1.1755,
      "step": 6934
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.879303961495744e-06,
      "loss": 1.3358,
      "step": 6935
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.877823028507961e-06,
      "loss": 1.1793,
      "step": 6936
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.876342095520179e-06,
      "loss": 1.4058,
      "step": 6937
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.874861162532396e-06,
      "loss": 1.5506,
      "step": 6938
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.873380229544614e-06,
      "loss": 1.3081,
      "step": 6939
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.871899296556831e-06,
      "loss": 1.3255,
      "step": 6940
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.87041836356905e-06,
      "loss": 1.3678,
      "step": 6941
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.868937430581266e-06,
      "loss": 1.4145,
      "step": 6942
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.867456497593486e-06,
      "loss": 1.3583,
      "step": 6943
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.865975564605701e-06,
      "loss": 1.147,
      "step": 6944
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.86449463161792e-06,
      "loss": 1.327,
      "step": 6945
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.863013698630138e-06,
      "loss": 1.3954,
      "step": 6946
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.861532765642356e-06,
      "loss": 1.2306,
      "step": 6947
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.860051832654573e-06,
      "loss": 1.3319,
      "step": 6948
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.85857089966679e-06,
      "loss": 1.2815,
      "step": 6949
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.857089966679008e-06,
      "loss": 1.2594,
      "step": 6950
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.855609033691226e-06,
      "loss": 1.0616,
      "step": 6951
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.854128100703443e-06,
      "loss": 1.1973,
      "step": 6952
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.852647167715662e-06,
      "loss": 1.2754,
      "step": 6953
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.85116623472788e-06,
      "loss": 1.1863,
      "step": 6954
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.849685301740097e-06,
      "loss": 1.3492,
      "step": 6955
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.848204368752315e-06,
      "loss": 1.4171,
      "step": 6956
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.846723435764532e-06,
      "loss": 1.3846,
      "step": 6957
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.84524250277675e-06,
      "loss": 1.1134,
      "step": 6958
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.843761569788967e-06,
      "loss": 1.7811,
      "step": 6959
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.842280636801185e-06,
      "loss": 1.2072,
      "step": 6960
    },
    {
      "epoch": 1.53,
      "learning_rate": 9.840799703813403e-06,
      "loss": 1.1763,
      "step": 6961
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.839318770825622e-06,
      "loss": 1.4938,
      "step": 6962
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.83783783783784e-06,
      "loss": 1.2856,
      "step": 6963
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.836356904850057e-06,
      "loss": 1.3262,
      "step": 6964
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.834875971862274e-06,
      "loss": 1.4086,
      "step": 6965
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.833395038874492e-06,
      "loss": 1.3587,
      "step": 6966
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.83191410588671e-06,
      "loss": 1.192,
      "step": 6967
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.830433172898927e-06,
      "loss": 1.4007,
      "step": 6968
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.828952239911144e-06,
      "loss": 1.277,
      "step": 6969
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.827471306923364e-06,
      "loss": 1.3181,
      "step": 6970
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.82599037393558e-06,
      "loss": 1.2571,
      "step": 6971
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.824509440947799e-06,
      "loss": 1.1082,
      "step": 6972
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.823028507960014e-06,
      "loss": 1.332,
      "step": 6973
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.821547574972234e-06,
      "loss": 1.2477,
      "step": 6974
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.820066641984451e-06,
      "loss": 1.351,
      "step": 6975
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.818585708996669e-06,
      "loss": 1.2597,
      "step": 6976
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.817104776008886e-06,
      "loss": 1.4148,
      "step": 6977
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.815623843021104e-06,
      "loss": 1.6095,
      "step": 6978
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.814142910033321e-06,
      "loss": 1.3675,
      "step": 6979
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.81266197704554e-06,
      "loss": 1.2658,
      "step": 6980
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.811181044057756e-06,
      "loss": 1.4318,
      "step": 6981
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.809700111069975e-06,
      "loss": 1.2073,
      "step": 6982
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.808219178082193e-06,
      "loss": 1.1732,
      "step": 6983
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.80673824509441e-06,
      "loss": 1.2386,
      "step": 6984
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.805257312106628e-06,
      "loss": 1.286,
      "step": 6985
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.803776379118846e-06,
      "loss": 1.4418,
      "step": 6986
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.802295446131063e-06,
      "loss": 1.2517,
      "step": 6987
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.80081451314328e-06,
      "loss": 1.1285,
      "step": 6988
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.7993335801555e-06,
      "loss": 1.5794,
      "step": 6989
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.797852647167716e-06,
      "loss": 1.2672,
      "step": 6990
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.796371714179935e-06,
      "loss": 1.5723,
      "step": 6991
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.794890781192152e-06,
      "loss": 1.1091,
      "step": 6992
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.79340984820437e-06,
      "loss": 1.5045,
      "step": 6993
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.791928915216587e-06,
      "loss": 1.2889,
      "step": 6994
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.790447982228805e-06,
      "loss": 1.3239,
      "step": 6995
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.788967049241022e-06,
      "loss": 1.1382,
      "step": 6996
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.787486116253242e-06,
      "loss": 1.2152,
      "step": 6997
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.786005183265457e-06,
      "loss": 1.1873,
      "step": 6998
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.784524250277677e-06,
      "loss": 1.3376,
      "step": 6999
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.783043317289892e-06,
      "loss": 1.383,
      "step": 7000
    },
    {
      "epoch": 1.54,
      "eval_loss": NaN,
      "eval_runtime": 4.3678,
      "eval_samples_per_second": 353.497,
      "eval_steps_per_second": 44.187,
      "step": 7000
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.781562384302112e-06,
      "loss": 1.2458,
      "step": 7001
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.780081451314327e-06,
      "loss": 1.2932,
      "step": 7002
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.778600518326547e-06,
      "loss": 1.3975,
      "step": 7003
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.777119585338764e-06,
      "loss": 1.1381,
      "step": 7004
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.775638652350982e-06,
      "loss": 1.4442,
      "step": 7005
    },
    {
      "epoch": 1.54,
      "learning_rate": 9.7741577193632e-06,
      "loss": 1.3119,
      "step": 7006
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.772676786375417e-06,
      "loss": 1.3905,
      "step": 7007
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.771195853387634e-06,
      "loss": 1.2714,
      "step": 7008
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.769714920399853e-06,
      "loss": 1.4078,
      "step": 7009
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.768233987412071e-06,
      "loss": 1.3733,
      "step": 7010
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.766753054424289e-06,
      "loss": 1.2614,
      "step": 7011
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.765272121436506e-06,
      "loss": 1.2966,
      "step": 7012
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.763791188448724e-06,
      "loss": 1.3334,
      "step": 7013
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.762310255460941e-06,
      "loss": 1.2298,
      "step": 7014
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.760829322473159e-06,
      "loss": 1.3352,
      "step": 7015
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.759348389485376e-06,
      "loss": 1.2664,
      "step": 7016
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.757867456497594e-06,
      "loss": 1.117,
      "step": 7017
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.756386523509813e-06,
      "loss": 1.3114,
      "step": 7018
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.754905590522029e-06,
      "loss": 1.3623,
      "step": 7019
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.753424657534248e-06,
      "loss": 1.2576,
      "step": 7020
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.751943724546465e-06,
      "loss": 1.2385,
      "step": 7021
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.750462791558683e-06,
      "loss": 1.2613,
      "step": 7022
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.7489818585709e-06,
      "loss": 1.3292,
      "step": 7023
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.747500925583118e-06,
      "loss": 1.3102,
      "step": 7024
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.746019992595335e-06,
      "loss": 1.6857,
      "step": 7025
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.744539059607555e-06,
      "loss": 1.4959,
      "step": 7026
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.74305812661977e-06,
      "loss": 1.3036,
      "step": 7027
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.74157719363199e-06,
      "loss": 1.2313,
      "step": 7028
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.740096260644206e-06,
      "loss": 1.246,
      "step": 7029
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.738615327656425e-06,
      "loss": 1.1782,
      "step": 7030
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.737134394668642e-06,
      "loss": 1.326,
      "step": 7031
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.73565346168086e-06,
      "loss": 1.2414,
      "step": 7032
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.734172528693077e-06,
      "loss": 1.2156,
      "step": 7033
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.732691595705295e-06,
      "loss": 1.4335,
      "step": 7034
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.731210662717512e-06,
      "loss": 1.2094,
      "step": 7035
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.729729729729732e-06,
      "loss": 1.4079,
      "step": 7036
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.728248796741947e-06,
      "loss": 1.4286,
      "step": 7037
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.726767863754167e-06,
      "loss": 1.3977,
      "step": 7038
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.725286930766384e-06,
      "loss": 1.2725,
      "step": 7039
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.723805997778602e-06,
      "loss": 1.5903,
      "step": 7040
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.722325064790819e-06,
      "loss": 1.3421,
      "step": 7041
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.720844131803037e-06,
      "loss": 1.3229,
      "step": 7042
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.719363198815254e-06,
      "loss": 1.2974,
      "step": 7043
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.717882265827472e-06,
      "loss": 1.0992,
      "step": 7044
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.71640133283969e-06,
      "loss": 1.3387,
      "step": 7045
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.714920399851907e-06,
      "loss": 1.3549,
      "step": 7046
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.713439466864126e-06,
      "loss": 1.3769,
      "step": 7047
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.711958533876343e-06,
      "loss": 1.3141,
      "step": 7048
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.710477600888561e-06,
      "loss": 1.1813,
      "step": 7049
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.708996667900778e-06,
      "loss": 1.2201,
      "step": 7050
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.707515734912996e-06,
      "loss": 1.3861,
      "step": 7051
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.706034801925213e-06,
      "loss": 1.2865,
      "step": 7052
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.704553868937433e-06,
      "loss": 1.3585,
      "step": 7053
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.703072935949649e-06,
      "loss": 1.4883,
      "step": 7054
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.701592002961868e-06,
      "loss": 1.5334,
      "step": 7055
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.700111069974084e-06,
      "loss": 1.4046,
      "step": 7056
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.698630136986303e-06,
      "loss": 1.2454,
      "step": 7057
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.697149203998519e-06,
      "loss": 1.377,
      "step": 7058
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.695668271010738e-06,
      "loss": 1.3954,
      "step": 7059
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.694187338022955e-06,
      "loss": 1.3556,
      "step": 7060
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.692706405035173e-06,
      "loss": 1.3794,
      "step": 7061
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.69122547204739e-06,
      "loss": 1.1453,
      "step": 7062
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.689744539059608e-06,
      "loss": 1.3992,
      "step": 7063
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.688263606071825e-06,
      "loss": 1.231,
      "step": 7064
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.686782673084045e-06,
      "loss": 1.3759,
      "step": 7065
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.68530174009626e-06,
      "loss": 1.2059,
      "step": 7066
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.68382080710848e-06,
      "loss": 1.4716,
      "step": 7067
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.682339874120697e-06,
      "loss": 1.4288,
      "step": 7068
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.680858941132915e-06,
      "loss": 1.2983,
      "step": 7069
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.679378008145132e-06,
      "loss": 1.2159,
      "step": 7070
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.67789707515735e-06,
      "loss": 1.3259,
      "step": 7071
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.676416142169567e-06,
      "loss": 1.2119,
      "step": 7072
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.674935209181785e-06,
      "loss": 1.3145,
      "step": 7073
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.673454276194004e-06,
      "loss": 1.3489,
      "step": 7074
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.67197334320622e-06,
      "loss": 1.2688,
      "step": 7075
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.670492410218439e-06,
      "loss": 1.1158,
      "step": 7076
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.669011477230657e-06,
      "loss": 1.195,
      "step": 7077
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.667530544242874e-06,
      "loss": 1.4085,
      "step": 7078
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.666049611255092e-06,
      "loss": 1.6998,
      "step": 7079
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.664568678267309e-06,
      "loss": 1.1471,
      "step": 7080
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.663087745279527e-06,
      "loss": 1.28,
      "step": 7081
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.661606812291746e-06,
      "loss": 1.3382,
      "step": 7082
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.660125879303962e-06,
      "loss": 1.3289,
      "step": 7083
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.65864494631618e-06,
      "loss": 1.3787,
      "step": 7084
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.657164013328397e-06,
      "loss": 1.3369,
      "step": 7085
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.655683080340616e-06,
      "loss": 1.2993,
      "step": 7086
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.654202147352833e-06,
      "loss": 1.2124,
      "step": 7087
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.652721214365051e-06,
      "loss": 1.1889,
      "step": 7088
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.651240281377268e-06,
      "loss": 1.2585,
      "step": 7089
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.649759348389486e-06,
      "loss": 1.2519,
      "step": 7090
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.648278415401703e-06,
      "loss": 1.3649,
      "step": 7091
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.646797482413923e-06,
      "loss": 1.4899,
      "step": 7092
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.645316549426138e-06,
      "loss": 1.318,
      "step": 7093
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.643835616438358e-06,
      "loss": 1.3672,
      "step": 7094
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.642354683450575e-06,
      "loss": 1.3769,
      "step": 7095
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.640873750462793e-06,
      "loss": 1.4666,
      "step": 7096
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.63939281747501e-06,
      "loss": 1.3703,
      "step": 7097
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.637911884487228e-06,
      "loss": 1.1542,
      "step": 7098
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.636430951499445e-06,
      "loss": 1.3443,
      "step": 7099
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.634950018511663e-06,
      "loss": 1.4281,
      "step": 7100
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.63346908552388e-06,
      "loss": 1.3627,
      "step": 7101
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.631988152536098e-06,
      "loss": 1.284,
      "step": 7102
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.630507219548317e-06,
      "loss": 1.3675,
      "step": 7103
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.629026286560535e-06,
      "loss": 1.2989,
      "step": 7104
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.627545353572752e-06,
      "loss": 1.3793,
      "step": 7105
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.62606442058497e-06,
      "loss": 1.2992,
      "step": 7106
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.624583487597187e-06,
      "loss": 1.1208,
      "step": 7107
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.623102554609405e-06,
      "loss": 1.2957,
      "step": 7108
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.621621621621622e-06,
      "loss": 1.4085,
      "step": 7109
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.62014068863384e-06,
      "loss": 1.2806,
      "step": 7110
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.618659755646059e-06,
      "loss": 1.2575,
      "step": 7111
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.617178822658275e-06,
      "loss": 1.3681,
      "step": 7112
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.615697889670494e-06,
      "loss": 1.3455,
      "step": 7113
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.61421695668271e-06,
      "loss": 1.2955,
      "step": 7114
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.612736023694929e-06,
      "loss": 1.2784,
      "step": 7115
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.611255090707146e-06,
      "loss": 1.4869,
      "step": 7116
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.609774157719364e-06,
      "loss": 1.2857,
      "step": 7117
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.608293224731581e-06,
      "loss": 0.9811,
      "step": 7118
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.606812291743799e-06,
      "loss": 1.3385,
      "step": 7119
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.605331358756017e-06,
      "loss": 1.4639,
      "step": 7120
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.603850425768236e-06,
      "loss": 1.2825,
      "step": 7121
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.602369492780452e-06,
      "loss": 1.4258,
      "step": 7122
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.60088855979267e-06,
      "loss": 1.4307,
      "step": 7123
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.599407626804888e-06,
      "loss": 1.1747,
      "step": 7124
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.597926693817106e-06,
      "loss": 1.2509,
      "step": 7125
    },
    {
      "epoch": 1.57,
      "eval_loss": NaN,
      "eval_runtime": 4.3711,
      "eval_samples_per_second": 353.232,
      "eval_steps_per_second": 44.154,
      "step": 7125
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.596445760829323e-06,
      "loss": 1.2376,
      "step": 7126
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.59496482784154e-06,
      "loss": 1.1774,
      "step": 7127
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.593483894853758e-06,
      "loss": 1.3731,
      "step": 7128
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.592002961865976e-06,
      "loss": 1.3226,
      "step": 7129
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.590522028878193e-06,
      "loss": 1.3606,
      "step": 7130
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.589041095890411e-06,
      "loss": 1.1895,
      "step": 7131
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.58756016290263e-06,
      "loss": 1.2125,
      "step": 7132
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.586079229914848e-06,
      "loss": 1.5084,
      "step": 7133
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.584598296927065e-06,
      "loss": 1.4817,
      "step": 7134
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.583117363939283e-06,
      "loss": 1.2503,
      "step": 7135
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.5816364309515e-06,
      "loss": 1.2298,
      "step": 7136
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.580155497963718e-06,
      "loss": 1.2708,
      "step": 7137
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.578674564975937e-06,
      "loss": 1.2953,
      "step": 7138
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.577193631988153e-06,
      "loss": 1.4608,
      "step": 7139
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.575712699000372e-06,
      "loss": 1.3768,
      "step": 7140
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.574231766012588e-06,
      "loss": 1.2092,
      "step": 7141
    },
    {
      "epoch": 1.57,
      "learning_rate": 9.572750833024807e-06,
      "loss": 1.2067,
      "step": 7142
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.571269900037024e-06,
      "loss": 1.2507,
      "step": 7143
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.569788967049242e-06,
      "loss": 1.4442,
      "step": 7144
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.56830803406146e-06,
      "loss": 1.335,
      "step": 7145
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.566827101073677e-06,
      "loss": 1.3225,
      "step": 7146
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.565346168085895e-06,
      "loss": 1.2877,
      "step": 7147
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.563865235098112e-06,
      "loss": 1.1837,
      "step": 7148
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.56238430211033e-06,
      "loss": 1.4028,
      "step": 7149
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.560903369122549e-06,
      "loss": 1.1897,
      "step": 7150
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.559422436134765e-06,
      "loss": 1.4432,
      "step": 7151
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.557941503146984e-06,
      "loss": 1.5374,
      "step": 7152
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.556460570159201e-06,
      "loss": 1.3505,
      "step": 7153
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.554979637171419e-06,
      "loss": 1.3141,
      "step": 7154
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.553498704183636e-06,
      "loss": 1.1592,
      "step": 7155
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.552017771195854e-06,
      "loss": 1.5844,
      "step": 7156
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.550536838208071e-06,
      "loss": 1.5219,
      "step": 7157
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.549055905220289e-06,
      "loss": 1.2093,
      "step": 7158
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.547574972232508e-06,
      "loss": 1.269,
      "step": 7159
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.546094039244726e-06,
      "loss": 1.1992,
      "step": 7160
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.544613106256943e-06,
      "loss": 1.2687,
      "step": 7161
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.54313217326916e-06,
      "loss": 1.292,
      "step": 7162
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.541651240281378e-06,
      "loss": 1.4478,
      "step": 7163
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.540170307293596e-06,
      "loss": 1.3409,
      "step": 7164
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.538689374305813e-06,
      "loss": 1.4616,
      "step": 7165
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.53720844131803e-06,
      "loss": 1.027,
      "step": 7166
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.53572750833025e-06,
      "loss": 1.3796,
      "step": 7167
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.534246575342466e-06,
      "loss": 1.2769,
      "step": 7168
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.532765642354685e-06,
      "loss": 1.4096,
      "step": 7169
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.5312847093669e-06,
      "loss": 0.9838,
      "step": 7170
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.52980377637912e-06,
      "loss": 1.5072,
      "step": 7171
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.528322843391338e-06,
      "loss": 1.357,
      "step": 7172
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.526841910403555e-06,
      "loss": 1.282,
      "step": 7173
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.525360977415773e-06,
      "loss": 1.2335,
      "step": 7174
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.52388004442799e-06,
      "loss": 1.4366,
      "step": 7175
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.522399111440208e-06,
      "loss": 1.4934,
      "step": 7176
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.520918178452427e-06,
      "loss": 1.2806,
      "step": 7177
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.519437245464643e-06,
      "loss": 1.2367,
      "step": 7178
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.517956312476862e-06,
      "loss": 1.2968,
      "step": 7179
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.516475379489078e-06,
      "loss": 1.4682,
      "step": 7180
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.514994446501297e-06,
      "loss": 1.0792,
      "step": 7181
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.513513513513514e-06,
      "loss": 1.271,
      "step": 7182
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.512032580525732e-06,
      "loss": 1.2267,
      "step": 7183
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.51055164753795e-06,
      "loss": 1.2204,
      "step": 7184
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.509070714550167e-06,
      "loss": 1.2371,
      "step": 7185
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.507589781562384e-06,
      "loss": 1.3393,
      "step": 7186
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.506108848574602e-06,
      "loss": 1.2888,
      "step": 7187
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.504627915586821e-06,
      "loss": 1.6303,
      "step": 7188
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.503146982599039e-06,
      "loss": 1.4424,
      "step": 7189
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.501666049611256e-06,
      "loss": 1.2105,
      "step": 7190
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.500185116623474e-06,
      "loss": 1.3125,
      "step": 7191
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.498704183635691e-06,
      "loss": 1.1194,
      "step": 7192
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.497223250647909e-06,
      "loss": 1.1819,
      "step": 7193
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.495742317660126e-06,
      "loss": 1.3033,
      "step": 7194
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.494261384672344e-06,
      "loss": 1.379,
      "step": 7195
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.492780451684563e-06,
      "loss": 1.4699,
      "step": 7196
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.491299518696779e-06,
      "loss": 1.2486,
      "step": 7197
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.489818585708998e-06,
      "loss": 1.3529,
      "step": 7198
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.488337652721216e-06,
      "loss": 1.2439,
      "step": 7199
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.486856719733433e-06,
      "loss": 1.288,
      "step": 7200
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.48537578674565e-06,
      "loss": 1.4149,
      "step": 7201
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.483894853757868e-06,
      "loss": 1.2547,
      "step": 7202
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.482413920770086e-06,
      "loss": 1.1075,
      "step": 7203
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.480932987782303e-06,
      "loss": 1.3845,
      "step": 7204
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.47945205479452e-06,
      "loss": 1.4836,
      "step": 7205
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.47797112180674e-06,
      "loss": 1.3387,
      "step": 7206
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.476490188818956e-06,
      "loss": 1.0871,
      "step": 7207
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.475009255831175e-06,
      "loss": 1.4226,
      "step": 7208
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.473528322843392e-06,
      "loss": 1.4039,
      "step": 7209
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.47204738985561e-06,
      "loss": 1.0962,
      "step": 7210
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.470566456867827e-06,
      "loss": 1.1422,
      "step": 7211
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.469085523880045e-06,
      "loss": 1.177,
      "step": 7212
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.467604590892263e-06,
      "loss": 1.4397,
      "step": 7213
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.46612365790448e-06,
      "loss": 1.7087,
      "step": 7214
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.464642724916698e-06,
      "loss": 1.3646,
      "step": 7215
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.463161791928917e-06,
      "loss": 1.237,
      "step": 7216
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.461680858941134e-06,
      "loss": 1.2224,
      "step": 7217
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.460199925953352e-06,
      "loss": 1.2159,
      "step": 7218
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.45871899296557e-06,
      "loss": 1.31,
      "step": 7219
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.457238059977787e-06,
      "loss": 1.3579,
      "step": 7220
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.455757126990004e-06,
      "loss": 1.3543,
      "step": 7221
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.454276194002222e-06,
      "loss": 1.2049,
      "step": 7222
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.452795261014441e-06,
      "loss": 1.4817,
      "step": 7223
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.451314328026657e-06,
      "loss": 1.4936,
      "step": 7224
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.449833395038876e-06,
      "loss": 1.2846,
      "step": 7225
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.448352462051092e-06,
      "loss": 1.4686,
      "step": 7226
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.446871529063311e-06,
      "loss": 1.1862,
      "step": 7227
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.445390596075529e-06,
      "loss": 1.3016,
      "step": 7228
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.443909663087746e-06,
      "loss": 1.245,
      "step": 7229
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.442428730099964e-06,
      "loss": 1.3512,
      "step": 7230
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.440947797112181e-06,
      "loss": 1.3671,
      "step": 7231
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.439466864124399e-06,
      "loss": 1.1992,
      "step": 7232
    },
    {
      "epoch": 1.59,
      "learning_rate": 9.437985931136618e-06,
      "loss": 1.2084,
      "step": 7233
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.436504998148834e-06,
      "loss": 1.1941,
      "step": 7234
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.435024065161053e-06,
      "loss": 1.3086,
      "step": 7235
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.433543132173269e-06,
      "loss": 1.2205,
      "step": 7236
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.432062199185488e-06,
      "loss": 1.5025,
      "step": 7237
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.430581266197706e-06,
      "loss": 1.4744,
      "step": 7238
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.429100333209923e-06,
      "loss": 1.2733,
      "step": 7239
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.42761940022214e-06,
      "loss": 1.4063,
      "step": 7240
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.426138467234358e-06,
      "loss": 1.3853,
      "step": 7241
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.424657534246576e-06,
      "loss": 1.2828,
      "step": 7242
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.423176601258793e-06,
      "loss": 1.1459,
      "step": 7243
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.42169566827101e-06,
      "loss": 1.3447,
      "step": 7244
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.42021473528323e-06,
      "loss": 1.3117,
      "step": 7245
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.418733802295447e-06,
      "loss": 1.2313,
      "step": 7246
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.417252869307665e-06,
      "loss": 1.4952,
      "step": 7247
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.415771936319882e-06,
      "loss": 1.3106,
      "step": 7248
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.4142910033321e-06,
      "loss": 1.048,
      "step": 7249
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.412810070344317e-06,
      "loss": 1.54,
      "step": 7250
    },
    {
      "epoch": 1.6,
      "eval_loss": NaN,
      "eval_runtime": 4.3748,
      "eval_samples_per_second": 352.928,
      "eval_steps_per_second": 44.116,
      "step": 7250
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.411329137356535e-06,
      "loss": 1.2415,
      "step": 7251
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.409848204368754e-06,
      "loss": 1.4598,
      "step": 7252
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.40836727138097e-06,
      "loss": 1.3985,
      "step": 7253
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.40688633839319e-06,
      "loss": 1.502,
      "step": 7254
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.405405405405407e-06,
      "loss": 1.4419,
      "step": 7255
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.403924472417624e-06,
      "loss": 1.4217,
      "step": 7256
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.402443539429842e-06,
      "loss": 1.3349,
      "step": 7257
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.40096260644206e-06,
      "loss": 1.4281,
      "step": 7258
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.399481673454277e-06,
      "loss": 1.3653,
      "step": 7259
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.398000740466494e-06,
      "loss": 1.3473,
      "step": 7260
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.396519807478712e-06,
      "loss": 1.3768,
      "step": 7261
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.395038874490931e-06,
      "loss": 1.3786,
      "step": 7262
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.393557941503147e-06,
      "loss": 1.3781,
      "step": 7263
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.392077008515366e-06,
      "loss": 1.2591,
      "step": 7264
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.390596075527582e-06,
      "loss": 1.2673,
      "step": 7265
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.389115142539801e-06,
      "loss": 1.2983,
      "step": 7266
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.387634209552019e-06,
      "loss": 1.2668,
      "step": 7267
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.386153276564236e-06,
      "loss": 1.3684,
      "step": 7268
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.384672343576454e-06,
      "loss": 1.3009,
      "step": 7269
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.383191410588671e-06,
      "loss": 1.3699,
      "step": 7270
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.381710477600889e-06,
      "loss": 1.4819,
      "step": 7271
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.380229544613108e-06,
      "loss": 1.2064,
      "step": 7272
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.378748611625325e-06,
      "loss": 1.3407,
      "step": 7273
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.377267678637543e-06,
      "loss": 1.0688,
      "step": 7274
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.37578674564976e-06,
      "loss": 1.2325,
      "step": 7275
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.374305812661978e-06,
      "loss": 1.3155,
      "step": 7276
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.372824879674195e-06,
      "loss": 1.3341,
      "step": 7277
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.371343946686413e-06,
      "loss": 1.3565,
      "step": 7278
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.36986301369863e-06,
      "loss": 1.0939,
      "step": 7279
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.368382080710848e-06,
      "loss": 1.4427,
      "step": 7280
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.366901147723067e-06,
      "loss": 1.2757,
      "step": 7281
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.365420214735283e-06,
      "loss": 1.3495,
      "step": 7282
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.363939281747502e-06,
      "loss": 1.1389,
      "step": 7283
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.36245834875972e-06,
      "loss": 1.2487,
      "step": 7284
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.360977415771937e-06,
      "loss": 1.1221,
      "step": 7285
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.359496482784155e-06,
      "loss": 1.3237,
      "step": 7286
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.358015549796372e-06,
      "loss": 1.4662,
      "step": 7287
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.35653461680859e-06,
      "loss": 1.5277,
      "step": 7288
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.355053683820809e-06,
      "loss": 1.2152,
      "step": 7289
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.353572750833025e-06,
      "loss": 1.2492,
      "step": 7290
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.352091817845244e-06,
      "loss": 1.3637,
      "step": 7291
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.35061088485746e-06,
      "loss": 1.1482,
      "step": 7292
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.349129951869679e-06,
      "loss": 1.3829,
      "step": 7293
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.347649018881897e-06,
      "loss": 1.4206,
      "step": 7294
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.346168085894114e-06,
      "loss": 1.2256,
      "step": 7295
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.344687152906332e-06,
      "loss": 1.396,
      "step": 7296
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.34320621991855e-06,
      "loss": 1.3782,
      "step": 7297
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.341725286930767e-06,
      "loss": 1.1715,
      "step": 7298
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.340244353942984e-06,
      "loss": 1.2635,
      "step": 7299
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.338763420955202e-06,
      "loss": 1.3427,
      "step": 7300
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.337282487967421e-06,
      "loss": 1.3191,
      "step": 7301
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.335801554979638e-06,
      "loss": 1.1414,
      "step": 7302
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.334320621991856e-06,
      "loss": 1.3317,
      "step": 7303
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.332839689004073e-06,
      "loss": 1.3214,
      "step": 7304
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.331358756016291e-06,
      "loss": 1.1712,
      "step": 7305
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.329877823028509e-06,
      "loss": 1.2938,
      "step": 7306
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.328396890040726e-06,
      "loss": 1.3049,
      "step": 7307
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.326915957052944e-06,
      "loss": 1.5957,
      "step": 7308
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.325435024065161e-06,
      "loss": 1.3884,
      "step": 7309
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.32395409107738e-06,
      "loss": 1.4333,
      "step": 7310
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.322473158089598e-06,
      "loss": 1.2279,
      "step": 7311
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.320992225101815e-06,
      "loss": 1.256,
      "step": 7312
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.319511292114033e-06,
      "loss": 1.1891,
      "step": 7313
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.31803035912625e-06,
      "loss": 1.2951,
      "step": 7314
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.316549426138468e-06,
      "loss": 1.3205,
      "step": 7315
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.315068493150685e-06,
      "loss": 1.4909,
      "step": 7316
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.313587560162903e-06,
      "loss": 1.2693,
      "step": 7317
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.312106627175122e-06,
      "loss": 1.4994,
      "step": 7318
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.310625694187338e-06,
      "loss": 1.3585,
      "step": 7319
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.309144761199557e-06,
      "loss": 1.1619,
      "step": 7320
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.307663828211773e-06,
      "loss": 1.1428,
      "step": 7321
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.306182895223992e-06,
      "loss": 1.4651,
      "step": 7322
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.30470196223621e-06,
      "loss": 1.2299,
      "step": 7323
    },
    {
      "epoch": 1.61,
      "learning_rate": 9.303221029248427e-06,
      "loss": 1.1233,
      "step": 7324
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.301740096260645e-06,
      "loss": 1.367,
      "step": 7325
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.300259163272862e-06,
      "loss": 1.0847,
      "step": 7326
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.29877823028508e-06,
      "loss": 1.4579,
      "step": 7327
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.297297297297299e-06,
      "loss": 1.3829,
      "step": 7328
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.295816364309515e-06,
      "loss": 1.5178,
      "step": 7329
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.294335431321734e-06,
      "loss": 1.4476,
      "step": 7330
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.292854498333952e-06,
      "loss": 1.1607,
      "step": 7331
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.291373565346169e-06,
      "loss": 1.6106,
      "step": 7332
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.289892632358387e-06,
      "loss": 1.3855,
      "step": 7333
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.288411699370604e-06,
      "loss": 1.1746,
      "step": 7334
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.286930766382822e-06,
      "loss": 1.2423,
      "step": 7335
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.285449833395039e-06,
      "loss": 1.232,
      "step": 7336
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.283968900407258e-06,
      "loss": 1.2079,
      "step": 7337
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.282487967419474e-06,
      "loss": 1.2795,
      "step": 7338
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.281007034431693e-06,
      "loss": 1.5215,
      "step": 7339
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.279526101443911e-06,
      "loss": 1.2642,
      "step": 7340
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.278045168456128e-06,
      "loss": 1.2492,
      "step": 7341
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.276564235468346e-06,
      "loss": 1.3927,
      "step": 7342
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.275083302480563e-06,
      "loss": 1.0797,
      "step": 7343
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.273602369492781e-06,
      "loss": 1.1336,
      "step": 7344
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.272121436505e-06,
      "loss": 1.0994,
      "step": 7345
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.270640503517216e-06,
      "loss": 1.0955,
      "step": 7346
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.269159570529435e-06,
      "loss": 1.3843,
      "step": 7347
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.267678637541651e-06,
      "loss": 1.4481,
      "step": 7348
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.26619770455387e-06,
      "loss": 1.3484,
      "step": 7349
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.264716771566086e-06,
      "loss": 1.2763,
      "step": 7350
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.263235838578305e-06,
      "loss": 1.085,
      "step": 7351
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.261754905590523e-06,
      "loss": 1.4047,
      "step": 7352
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.26027397260274e-06,
      "loss": 1.3875,
      "step": 7353
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.258793039614958e-06,
      "loss": 1.0729,
      "step": 7354
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.257312106627175e-06,
      "loss": 1.1317,
      "step": 7355
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.255831173639393e-06,
      "loss": 1.1025,
      "step": 7356
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.254350240651612e-06,
      "loss": 1.4932,
      "step": 7357
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.25286930766383e-06,
      "loss": 1.4086,
      "step": 7358
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.251388374676047e-06,
      "loss": 1.2767,
      "step": 7359
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.249907441688265e-06,
      "loss": 1.3045,
      "step": 7360
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.248426508700482e-06,
      "loss": 1.4394,
      "step": 7361
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.2469455757127e-06,
      "loss": 1.4404,
      "step": 7362
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.245464642724917e-06,
      "loss": 1.2994,
      "step": 7363
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.243983709737135e-06,
      "loss": 1.2837,
      "step": 7364
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.242502776749352e-06,
      "loss": 1.4048,
      "step": 7365
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.241021843761571e-06,
      "loss": 1.1393,
      "step": 7366
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.239540910773789e-06,
      "loss": 1.3792,
      "step": 7367
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.238059977786006e-06,
      "loss": 1.5113,
      "step": 7368
    },
    {
      "epoch": 1.62,
      "learning_rate": 9.236579044798224e-06,
      "loss": 1.3398,
      "step": 7369
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.235098111810441e-06,
      "loss": 1.3997,
      "step": 7370
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.233617178822659e-06,
      "loss": 1.328,
      "step": 7371
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.232136245834877e-06,
      "loss": 1.4792,
      "step": 7372
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.230655312847094e-06,
      "loss": 1.5051,
      "step": 7373
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.229174379859313e-06,
      "loss": 1.2422,
      "step": 7374
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.227693446871529e-06,
      "loss": 1.1274,
      "step": 7375
    },
    {
      "epoch": 1.63,
      "eval_loss": NaN,
      "eval_runtime": 4.3833,
      "eval_samples_per_second": 352.247,
      "eval_steps_per_second": 44.031,
      "step": 7375
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.226212513883748e-06,
      "loss": 1.2505,
      "step": 7376
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.224731580895964e-06,
      "loss": 1.3816,
      "step": 7377
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.223250647908183e-06,
      "loss": 1.2482,
      "step": 7378
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.2217697149204e-06,
      "loss": 1.3252,
      "step": 7379
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.220288781932618e-06,
      "loss": 1.5296,
      "step": 7380
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.218807848944836e-06,
      "loss": 1.4374,
      "step": 7381
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.217326915957053e-06,
      "loss": 1.4842,
      "step": 7382
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.215845982969271e-06,
      "loss": 1.1714,
      "step": 7383
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.21436504998149e-06,
      "loss": 1.3069,
      "step": 7384
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.212884116993706e-06,
      "loss": 1.2375,
      "step": 7385
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.211403184005925e-06,
      "loss": 1.1238,
      "step": 7386
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.209922251018143e-06,
      "loss": 1.4644,
      "step": 7387
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.20844131803036e-06,
      "loss": 1.2829,
      "step": 7388
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.206960385042578e-06,
      "loss": 1.2282,
      "step": 7389
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.205479452054795e-06,
      "loss": 1.2757,
      "step": 7390
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.203998519067013e-06,
      "loss": 1.346,
      "step": 7391
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.20251758607923e-06,
      "loss": 1.3721,
      "step": 7392
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.201036653091448e-06,
      "loss": 1.351,
      "step": 7393
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.199555720103665e-06,
      "loss": 1.4885,
      "step": 7394
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.198074787115884e-06,
      "loss": 1.2155,
      "step": 7395
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.196593854128102e-06,
      "loss": 1.1579,
      "step": 7396
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.19511292114032e-06,
      "loss": 1.2696,
      "step": 7397
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.193631988152537e-06,
      "loss": 1.1605,
      "step": 7398
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.192151055164755e-06,
      "loss": 1.2496,
      "step": 7399
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.190670122176972e-06,
      "loss": 1.4258,
      "step": 7400
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.189189189189191e-06,
      "loss": 1.3521,
      "step": 7401
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.187708256201407e-06,
      "loss": 1.5012,
      "step": 7402
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.186227323213626e-06,
      "loss": 1.3115,
      "step": 7403
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.184746390225842e-06,
      "loss": 1.5962,
      "step": 7404
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.183265457238061e-06,
      "loss": 1.4087,
      "step": 7405
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.181784524250277e-06,
      "loss": 1.3138,
      "step": 7406
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.180303591262496e-06,
      "loss": 1.5884,
      "step": 7407
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.178822658274714e-06,
      "loss": 1.2975,
      "step": 7408
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.177341725286931e-06,
      "loss": 1.4767,
      "step": 7409
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.175860792299149e-06,
      "loss": 1.6817,
      "step": 7410
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.174379859311366e-06,
      "loss": 1.4882,
      "step": 7411
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.172898926323584e-06,
      "loss": 1.3618,
      "step": 7412
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.171417993335803e-06,
      "loss": 1.2736,
      "step": 7413
    },
    {
      "epoch": 1.63,
      "learning_rate": 9.169937060348019e-06,
      "loss": 1.3516,
      "step": 7414
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.168456127360238e-06,
      "loss": 1.5289,
      "step": 7415
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.166975194372456e-06,
      "loss": 1.2018,
      "step": 7416
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.165494261384673e-06,
      "loss": 1.2746,
      "step": 7417
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.16401332839689e-06,
      "loss": 1.3974,
      "step": 7418
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.162532395409108e-06,
      "loss": 1.3151,
      "step": 7419
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.161051462421326e-06,
      "loss": 1.3097,
      "step": 7420
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.159570529433543e-06,
      "loss": 1.2991,
      "step": 7421
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.158089596445763e-06,
      "loss": 1.1348,
      "step": 7422
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.15660866345798e-06,
      "loss": 1.3114,
      "step": 7423
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.155127730470198e-06,
      "loss": 1.3814,
      "step": 7424
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.153646797482415e-06,
      "loss": 1.5085,
      "step": 7425
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.152165864494633e-06,
      "loss": 1.5287,
      "step": 7426
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.15068493150685e-06,
      "loss": 1.374,
      "step": 7427
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.149203998519068e-06,
      "loss": 1.0808,
      "step": 7428
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.147723065531285e-06,
      "loss": 1.2924,
      "step": 7429
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.146242132543504e-06,
      "loss": 1.127,
      "step": 7430
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.14476119955572e-06,
      "loss": 1.2711,
      "step": 7431
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.14328026656794e-06,
      "loss": 1.126,
      "step": 7432
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.141799333580155e-06,
      "loss": 1.4545,
      "step": 7433
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.140318400592374e-06,
      "loss": 1.5602,
      "step": 7434
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.138837467604592e-06,
      "loss": 1.2902,
      "step": 7435
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.13735653461681e-06,
      "loss": 1.382,
      "step": 7436
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.135875601629027e-06,
      "loss": 1.2751,
      "step": 7437
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.134394668641244e-06,
      "loss": 1.3252,
      "step": 7438
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.132913735653462e-06,
      "loss": 1.33,
      "step": 7439
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.131432802665681e-06,
      "loss": 1.2932,
      "step": 7440
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.129951869677897e-06,
      "loss": 1.1542,
      "step": 7441
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.128470936690116e-06,
      "loss": 1.3617,
      "step": 7442
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.126990003702332e-06,
      "loss": 1.3781,
      "step": 7443
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.125509070714551e-06,
      "loss": 1.3848,
      "step": 7444
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.124028137726769e-06,
      "loss": 1.4256,
      "step": 7445
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.122547204738986e-06,
      "loss": 1.3608,
      "step": 7446
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.121066271751204e-06,
      "loss": 1.296,
      "step": 7447
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.119585338763421e-06,
      "loss": 1.2137,
      "step": 7448
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.118104405775639e-06,
      "loss": 1.3568,
      "step": 7449
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.116623472787856e-06,
      "loss": 1.4507,
      "step": 7450
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.115142539800076e-06,
      "loss": 1.2806,
      "step": 7451
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.113661606812293e-06,
      "loss": 1.1145,
      "step": 7452
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.11218067382451e-06,
      "loss": 1.2647,
      "step": 7453
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.110699740836728e-06,
      "loss": 1.4051,
      "step": 7454
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.109218807848946e-06,
      "loss": 1.2774,
      "step": 7455
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.107737874861163e-06,
      "loss": 1.2652,
      "step": 7456
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.10625694187338e-06,
      "loss": 1.4253,
      "step": 7457
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.104776008885598e-06,
      "loss": 1.1698,
      "step": 7458
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.103295075897817e-06,
      "loss": 1.2327,
      "step": 7459
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.101814142910033e-06,
      "loss": 1.2671,
      "step": 7460
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.100333209922252e-06,
      "loss": 1.2977,
      "step": 7461
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.098852276934468e-06,
      "loss": 1.4767,
      "step": 7462
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.097371343946687e-06,
      "loss": 1.2847,
      "step": 7463
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.095890410958905e-06,
      "loss": 1.4353,
      "step": 7464
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.094409477971123e-06,
      "loss": 1.214,
      "step": 7465
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.09292854498334e-06,
      "loss": 1.3557,
      "step": 7466
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.091447611995558e-06,
      "loss": 1.2173,
      "step": 7467
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.089966679007775e-06,
      "loss": 1.403,
      "step": 7468
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.088485746019994e-06,
      "loss": 1.0907,
      "step": 7469
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.08700481303221e-06,
      "loss": 1.2812,
      "step": 7470
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.08552388004443e-06,
      "loss": 1.5983,
      "step": 7471
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.084042947056647e-06,
      "loss": 1.3337,
      "step": 7472
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.082562014068864e-06,
      "loss": 1.2057,
      "step": 7473
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.081081081081082e-06,
      "loss": 1.2602,
      "step": 7474
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.0796001480933e-06,
      "loss": 1.6218,
      "step": 7475
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.078119215105517e-06,
      "loss": 1.2353,
      "step": 7476
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.076638282117734e-06,
      "loss": 1.3986,
      "step": 7477
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.075157349129952e-06,
      "loss": 1.1709,
      "step": 7478
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.07367641614217e-06,
      "loss": 1.3615,
      "step": 7479
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.072195483154389e-06,
      "loss": 1.3065,
      "step": 7480
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.070714550166606e-06,
      "loss": 1.2518,
      "step": 7481
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.069233617178824e-06,
      "loss": 1.1994,
      "step": 7482
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.067752684191041e-06,
      "loss": 1.2023,
      "step": 7483
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.066271751203259e-06,
      "loss": 1.1444,
      "step": 7484
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.064790818215476e-06,
      "loss": 1.8485,
      "step": 7485
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.063309885227695e-06,
      "loss": 1.3585,
      "step": 7486
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.061828952239911e-06,
      "loss": 1.2993,
      "step": 7487
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.06034801925213e-06,
      "loss": 0.9658,
      "step": 7488
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.058867086264346e-06,
      "loss": 1.3312,
      "step": 7489
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.057386153276566e-06,
      "loss": 1.3061,
      "step": 7490
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.055905220288783e-06,
      "loss": 1.4241,
      "step": 7491
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.054424287301e-06,
      "loss": 1.1648,
      "step": 7492
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.052943354313218e-06,
      "loss": 1.399,
      "step": 7493
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.051462421325436e-06,
      "loss": 1.274,
      "step": 7494
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.049981488337653e-06,
      "loss": 1.3386,
      "step": 7495
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.048500555349872e-06,
      "loss": 1.3536,
      "step": 7496
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.047019622362088e-06,
      "loss": 1.344,
      "step": 7497
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.045538689374307e-06,
      "loss": 1.347,
      "step": 7498
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.044057756386523e-06,
      "loss": 1.57,
      "step": 7499
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.042576823398742e-06,
      "loss": 1.3435,
      "step": 7500
    },
    {
      "epoch": 1.65,
      "eval_loss": NaN,
      "eval_runtime": 4.3759,
      "eval_samples_per_second": 352.845,
      "eval_steps_per_second": 44.106,
      "step": 7500
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.04109589041096e-06,
      "loss": 1.1931,
      "step": 7501
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.039614957423177e-06,
      "loss": 1.417,
      "step": 7502
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.038134024435395e-06,
      "loss": 1.1806,
      "step": 7503
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.036653091447612e-06,
      "loss": 1.2799,
      "step": 7504
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.03517215845983e-06,
      "loss": 1.3208,
      "step": 7505
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.033691225472047e-06,
      "loss": 1.4162,
      "step": 7506
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.032210292484265e-06,
      "loss": 1.1162,
      "step": 7507
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.030729359496484e-06,
      "loss": 1.2041,
      "step": 7508
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.029248426508702e-06,
      "loss": 1.2225,
      "step": 7509
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.02776749352092e-06,
      "loss": 1.3287,
      "step": 7510
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.026286560533137e-06,
      "loss": 1.3222,
      "step": 7511
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.024805627545354e-06,
      "loss": 1.1173,
      "step": 7512
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.023324694557572e-06,
      "loss": 1.3871,
      "step": 7513
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.02184376156979e-06,
      "loss": 0.9615,
      "step": 7514
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.020362828582009e-06,
      "loss": 1.1698,
      "step": 7515
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.018881895594224e-06,
      "loss": 1.4777,
      "step": 7516
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.017400962606444e-06,
      "loss": 1.3898,
      "step": 7517
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.01592002961866e-06,
      "loss": 1.2056,
      "step": 7518
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.014439096630879e-06,
      "loss": 1.3444,
      "step": 7519
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.012958163643096e-06,
      "loss": 1.3396,
      "step": 7520
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.011477230655314e-06,
      "loss": 1.2901,
      "step": 7521
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.009996297667531e-06,
      "loss": 1.4194,
      "step": 7522
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.008515364679749e-06,
      "loss": 1.2416,
      "step": 7523
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.007034431691966e-06,
      "loss": 1.2169,
      "step": 7524
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.005553498704185e-06,
      "loss": 1.3302,
      "step": 7525
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.004072565716401e-06,
      "loss": 1.275,
      "step": 7526
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.00259163272862e-06,
      "loss": 0.9564,
      "step": 7527
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.001110699740836e-06,
      "loss": 1.5599,
      "step": 7528
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.999629766753055e-06,
      "loss": 1.2753,
      "step": 7529
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.998148833765273e-06,
      "loss": 1.4378,
      "step": 7530
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.99666790077749e-06,
      "loss": 1.1359,
      "step": 7531
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.995186967789708e-06,
      "loss": 1.2112,
      "step": 7532
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.993706034801926e-06,
      "loss": 1.3562,
      "step": 7533
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.992225101814143e-06,
      "loss": 1.2158,
      "step": 7534
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.99074416882636e-06,
      "loss": 1.391,
      "step": 7535
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.98926323583858e-06,
      "loss": 1.1603,
      "step": 7536
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.987782302850797e-06,
      "loss": 1.4026,
      "step": 7537
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.986301369863015e-06,
      "loss": 1.2394,
      "step": 7538
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.984820436875232e-06,
      "loss": 1.5003,
      "step": 7539
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.98333950388745e-06,
      "loss": 1.4086,
      "step": 7540
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.981858570899667e-06,
      "loss": 1.3143,
      "step": 7541
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.980377637911885e-06,
      "loss": 1.348,
      "step": 7542
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.978896704924102e-06,
      "loss": 1.3431,
      "step": 7543
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.977415771936322e-06,
      "loss": 1.2485,
      "step": 7544
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.975934838948537e-06,
      "loss": 1.4025,
      "step": 7545
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.974453905960757e-06,
      "loss": 1.4428,
      "step": 7546
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.972972972972974e-06,
      "loss": 1.5188,
      "step": 7547
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.971492039985192e-06,
      "loss": 1.3463,
      "step": 7548
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.97001110699741e-06,
      "loss": 1.5153,
      "step": 7549
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.968530174009627e-06,
      "loss": 1.2706,
      "step": 7550
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.967049241021844e-06,
      "loss": 1.302,
      "step": 7551
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.965568308034062e-06,
      "loss": 1.3848,
      "step": 7552
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.96408737504628e-06,
      "loss": 1.1406,
      "step": 7553
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.962606442058498e-06,
      "loss": 1.3178,
      "step": 7554
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.961125509070714e-06,
      "loss": 1.4267,
      "step": 7555
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.959644576082933e-06,
      "loss": 1.2551,
      "step": 7556
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.958163643095151e-06,
      "loss": 1.0842,
      "step": 7557
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.956682710107369e-06,
      "loss": 1.2594,
      "step": 7558
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.955201777119586e-06,
      "loss": 1.3885,
      "step": 7559
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.953720844131804e-06,
      "loss": 1.1004,
      "step": 7560
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.952239911144021e-06,
      "loss": 1.348,
      "step": 7561
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.950758978156239e-06,
      "loss": 1.2842,
      "step": 7562
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.949278045168456e-06,
      "loss": 1.0631,
      "step": 7563
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.947797112180675e-06,
      "loss": 1.3606,
      "step": 7564
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.946316179192893e-06,
      "loss": 1.2395,
      "step": 7565
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.94483524620511e-06,
      "loss": 1.4337,
      "step": 7566
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.943354313217328e-06,
      "loss": 1.6041,
      "step": 7567
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.941873380229545e-06,
      "loss": 1.3127,
      "step": 7568
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.940392447241763e-06,
      "loss": 1.2792,
      "step": 7569
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.93891151425398e-06,
      "loss": 1.3212,
      "step": 7570
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.937430581266198e-06,
      "loss": 1.4305,
      "step": 7571
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.935949648278415e-06,
      "loss": 1.4184,
      "step": 7572
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.934468715290635e-06,
      "loss": 1.2495,
      "step": 7573
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.93298778230285e-06,
      "loss": 1.3019,
      "step": 7574
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.93150684931507e-06,
      "loss": 1.1099,
      "step": 7575
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.930025916327287e-06,
      "loss": 1.2021,
      "step": 7576
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.928544983339505e-06,
      "loss": 1.4686,
      "step": 7577
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.927064050351722e-06,
      "loss": 1.2791,
      "step": 7578
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.92558311736394e-06,
      "loss": 1.2908,
      "step": 7579
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.924102184376157e-06,
      "loss": 1.3587,
      "step": 7580
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.922621251388376e-06,
      "loss": 1.4082,
      "step": 7581
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.921140318400592e-06,
      "loss": 1.2974,
      "step": 7582
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.919659385412812e-06,
      "loss": 1.1413,
      "step": 7583
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.918178452425027e-06,
      "loss": 1.1562,
      "step": 7584
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.916697519437247e-06,
      "loss": 1.2973,
      "step": 7585
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.915216586449464e-06,
      "loss": 1.4859,
      "step": 7586
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.913735653461682e-06,
      "loss": 1.2813,
      "step": 7587
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.912254720473899e-06,
      "loss": 1.2191,
      "step": 7588
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.910773787486117e-06,
      "loss": 1.1326,
      "step": 7589
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.909292854498334e-06,
      "loss": 1.1393,
      "step": 7590
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.907811921510552e-06,
      "loss": 1.383,
      "step": 7591
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.90633098852277e-06,
      "loss": 1.2221,
      "step": 7592
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.904850055534988e-06,
      "loss": 1.1488,
      "step": 7593
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.903369122547206e-06,
      "loss": 1.2786,
      "step": 7594
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.901888189559423e-06,
      "loss": 1.3154,
      "step": 7595
    },
    {
      "epoch": 1.67,
      "learning_rate": 8.900407256571641e-06,
      "loss": 1.3175,
      "step": 7596
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.898926323583858e-06,
      "loss": 1.5374,
      "step": 7597
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.897445390596076e-06,
      "loss": 1.4526,
      "step": 7598
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.895964457608293e-06,
      "loss": 1.2665,
      "step": 7599
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.894483524620513e-06,
      "loss": 1.1191,
      "step": 7600
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.893002591632729e-06,
      "loss": 1.2878,
      "step": 7601
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.891521658644948e-06,
      "loss": 1.277,
      "step": 7602
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.890040725657165e-06,
      "loss": 1.2011,
      "step": 7603
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.888559792669383e-06,
      "loss": 1.2203,
      "step": 7604
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.8870788596816e-06,
      "loss": 1.1953,
      "step": 7605
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.885597926693818e-06,
      "loss": 1.2462,
      "step": 7606
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.884116993706035e-06,
      "loss": 1.2786,
      "step": 7607
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.882636060718253e-06,
      "loss": 1.3465,
      "step": 7608
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.88115512773047e-06,
      "loss": 1.2368,
      "step": 7609
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.87967419474269e-06,
      "loss": 1.3816,
      "step": 7610
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.878193261754905e-06,
      "loss": 1.3756,
      "step": 7611
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.876712328767125e-06,
      "loss": 1.4185,
      "step": 7612
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.87523139577934e-06,
      "loss": 1.296,
      "step": 7613
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.87375046279156e-06,
      "loss": 1.3255,
      "step": 7614
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.872269529803777e-06,
      "loss": 1.2506,
      "step": 7615
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.870788596815995e-06,
      "loss": 1.2441,
      "step": 7616
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.869307663828212e-06,
      "loss": 1.132,
      "step": 7617
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.86782673084043e-06,
      "loss": 1.8814,
      "step": 7618
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.866345797852647e-06,
      "loss": 1.2633,
      "step": 7619
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.864864864864866e-06,
      "loss": 1.1033,
      "step": 7620
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.863383931877084e-06,
      "loss": 1.23,
      "step": 7621
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.861902998889301e-06,
      "loss": 1.405,
      "step": 7622
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.860422065901519e-06,
      "loss": 1.457,
      "step": 7623
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.858941132913737e-06,
      "loss": 1.3106,
      "step": 7624
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.857460199925954e-06,
      "loss": 1.2536,
      "step": 7625
    },
    {
      "epoch": 1.68,
      "eval_loss": NaN,
      "eval_runtime": 4.3736,
      "eval_samples_per_second": 353.029,
      "eval_steps_per_second": 44.129,
      "step": 7625
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.855979266938172e-06,
      "loss": 1.2336,
      "step": 7626
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.854498333950389e-06,
      "loss": 1.1432,
      "step": 7627
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.853017400962607e-06,
      "loss": 1.0543,
      "step": 7628
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.851536467974826e-06,
      "loss": 1.4803,
      "step": 7629
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.850055534987042e-06,
      "loss": 1.4176,
      "step": 7630
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.84857460199926e-06,
      "loss": 1.162,
      "step": 7631
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.847093669011478e-06,
      "loss": 1.257,
      "step": 7632
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.845612736023696e-06,
      "loss": 1.3338,
      "step": 7633
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.844131803035913e-06,
      "loss": 1.2338,
      "step": 7634
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.842650870048131e-06,
      "loss": 1.3774,
      "step": 7635
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.841169937060348e-06,
      "loss": 1.2934,
      "step": 7636
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.839689004072568e-06,
      "loss": 1.1835,
      "step": 7637
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.838208071084783e-06,
      "loss": 1.036,
      "step": 7638
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.836727138097003e-06,
      "loss": 1.3766,
      "step": 7639
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.835246205109218e-06,
      "loss": 1.1015,
      "step": 7640
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.833765272121438e-06,
      "loss": 1.3209,
      "step": 7641
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.832284339133655e-06,
      "loss": 1.2315,
      "step": 7642
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.830803406145873e-06,
      "loss": 1.1255,
      "step": 7643
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.82932247315809e-06,
      "loss": 1.4294,
      "step": 7644
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.827841540170308e-06,
      "loss": 1.3138,
      "step": 7645
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.826360607182525e-06,
      "loss": 1.2723,
      "step": 7646
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.824879674194743e-06,
      "loss": 1.1559,
      "step": 7647
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.82339874120696e-06,
      "loss": 1.3849,
      "step": 7648
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.82191780821918e-06,
      "loss": 1.2163,
      "step": 7649
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.820436875231397e-06,
      "loss": 1.2936,
      "step": 7650
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.818955942243615e-06,
      "loss": 1.2598,
      "step": 7651
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.817475009255832e-06,
      "loss": 1.4109,
      "step": 7652
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.81599407626805e-06,
      "loss": 1.1677,
      "step": 7653
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.814513143280267e-06,
      "loss": 1.4283,
      "step": 7654
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.813032210292485e-06,
      "loss": 1.3986,
      "step": 7655
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.811551277304702e-06,
      "loss": 1.1866,
      "step": 7656
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.81007034431692e-06,
      "loss": 1.3724,
      "step": 7657
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.808589411329139e-06,
      "loss": 1.2536,
      "step": 7658
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.807108478341356e-06,
      "loss": 1.448,
      "step": 7659
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.805627545353574e-06,
      "loss": 1.3942,
      "step": 7660
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.804146612365791e-06,
      "loss": 1.3577,
      "step": 7661
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.802665679378009e-06,
      "loss": 1.1208,
      "step": 7662
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.801184746390226e-06,
      "loss": 1.3251,
      "step": 7663
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.799703813402444e-06,
      "loss": 1.2621,
      "step": 7664
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.798222880414661e-06,
      "loss": 1.2583,
      "step": 7665
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.79674194742688e-06,
      "loss": 1.2202,
      "step": 7666
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.795261014439097e-06,
      "loss": 1.1967,
      "step": 7667
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.793780081451316e-06,
      "loss": 1.3357,
      "step": 7668
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.792299148463532e-06,
      "loss": 1.4037,
      "step": 7669
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.79081821547575e-06,
      "loss": 1.4604,
      "step": 7670
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.789337282487968e-06,
      "loss": 1.3103,
      "step": 7671
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.787856349500186e-06,
      "loss": 1.2693,
      "step": 7672
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.786375416512403e-06,
      "loss": 1.5234,
      "step": 7673
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.78489448352462e-06,
      "loss": 1.2247,
      "step": 7674
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.783413550536838e-06,
      "loss": 1.522,
      "step": 7675
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.781932617549058e-06,
      "loss": 1.4546,
      "step": 7676
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.780451684561273e-06,
      "loss": 1.2641,
      "step": 7677
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.778970751573493e-06,
      "loss": 1.1703,
      "step": 7678
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.77748981858571e-06,
      "loss": 1.1862,
      "step": 7679
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.776008885597928e-06,
      "loss": 1.3976,
      "step": 7680
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.774527952610145e-06,
      "loss": 1.1167,
      "step": 7681
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.773047019622363e-06,
      "loss": 1.1286,
      "step": 7682
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.77156608663458e-06,
      "loss": 1.3824,
      "step": 7683
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.770085153646798e-06,
      "loss": 1.194,
      "step": 7684
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.768604220659017e-06,
      "loss": 1.2073,
      "step": 7685
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.767123287671233e-06,
      "loss": 1.4243,
      "step": 7686
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.765642354683452e-06,
      "loss": 1.3566,
      "step": 7687
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.76416142169567e-06,
      "loss": 1.4659,
      "step": 7688
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.762680488707887e-06,
      "loss": 1.3949,
      "step": 7689
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.761199555720104e-06,
      "loss": 1.5529,
      "step": 7690
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.759718622732322e-06,
      "loss": 1.1832,
      "step": 7691
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.75823768974454e-06,
      "loss": 1.2136,
      "step": 7692
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.756756756756759e-06,
      "loss": 1.4889,
      "step": 7693
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.755275823768975e-06,
      "loss": 1.0909,
      "step": 7694
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.753794890781194e-06,
      "loss": 1.266,
      "step": 7695
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.75231395779341e-06,
      "loss": 1.2096,
      "step": 7696
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.750833024805629e-06,
      "loss": 1.1278,
      "step": 7697
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.749352091817846e-06,
      "loss": 1.1838,
      "step": 7698
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.747871158830064e-06,
      "loss": 1.2129,
      "step": 7699
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.746390225842281e-06,
      "loss": 1.2066,
      "step": 7700
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.744909292854499e-06,
      "loss": 1.2872,
      "step": 7701
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.743428359866716e-06,
      "loss": 1.469,
      "step": 7702
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.741947426878934e-06,
      "loss": 1.1181,
      "step": 7703
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.740466493891151e-06,
      "loss": 1.0699,
      "step": 7704
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.73898556090337e-06,
      "loss": 1.0434,
      "step": 7705
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.737504627915588e-06,
      "loss": 1.3057,
      "step": 7706
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.736023694927806e-06,
      "loss": 1.2401,
      "step": 7707
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.734542761940023e-06,
      "loss": 1.1755,
      "step": 7708
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.73306182895224e-06,
      "loss": 1.2155,
      "step": 7709
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.731580895964458e-06,
      "loss": 1.2187,
      "step": 7710
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.730099962976676e-06,
      "loss": 1.1729,
      "step": 7711
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.728619029988893e-06,
      "loss": 1.3461,
      "step": 7712
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.72713809700111e-06,
      "loss": 1.263,
      "step": 7713
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.72565716401333e-06,
      "loss": 1.1414,
      "step": 7714
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.724176231025547e-06,
      "loss": 1.3261,
      "step": 7715
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.722695298037765e-06,
      "loss": 1.2521,
      "step": 7716
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.721214365049983e-06,
      "loss": 1.3594,
      "step": 7717
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.7197334320622e-06,
      "loss": 1.2489,
      "step": 7718
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.718252499074418e-06,
      "loss": 1.2753,
      "step": 7719
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.716771566086635e-06,
      "loss": 1.3691,
      "step": 7720
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.715290633098853e-06,
      "loss": 1.3379,
      "step": 7721
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.713809700111072e-06,
      "loss": 1.2294,
      "step": 7722
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.712328767123288e-06,
      "loss": 1.6436,
      "step": 7723
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.710847834135507e-06,
      "loss": 1.2651,
      "step": 7724
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.709366901147723e-06,
      "loss": 1.3947,
      "step": 7725
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.707885968159942e-06,
      "loss": 1.3633,
      "step": 7726
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.70640503517216e-06,
      "loss": 1.3432,
      "step": 7727
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.704924102184377e-06,
      "loss": 1.1738,
      "step": 7728
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.703443169196594e-06,
      "loss": 1.3366,
      "step": 7729
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.701962236208812e-06,
      "loss": 1.2808,
      "step": 7730
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.70048130322103e-06,
      "loss": 1.2993,
      "step": 7731
    },
    {
      "epoch": 1.7,
      "learning_rate": 8.699000370233249e-06,
      "loss": 1.4342,
      "step": 7732
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.697519437245464e-06,
      "loss": 1.2553,
      "step": 7733
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.696038504257684e-06,
      "loss": 1.3339,
      "step": 7734
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.694557571269901e-06,
      "loss": 1.1722,
      "step": 7735
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.693076638282119e-06,
      "loss": 1.3116,
      "step": 7736
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.691595705294336e-06,
      "loss": 1.471,
      "step": 7737
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.690114772306554e-06,
      "loss": 1.4928,
      "step": 7738
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.688633839318771e-06,
      "loss": 1.5418,
      "step": 7739
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.687152906330989e-06,
      "loss": 1.1875,
      "step": 7740
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.685671973343206e-06,
      "loss": 1.3107,
      "step": 7741
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.684191040355424e-06,
      "loss": 1.1821,
      "step": 7742
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.682710107367643e-06,
      "loss": 1.1719,
      "step": 7743
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.68122917437986e-06,
      "loss": 1.1777,
      "step": 7744
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.679748241392078e-06,
      "loss": 1.205,
      "step": 7745
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.678267308404296e-06,
      "loss": 1.2754,
      "step": 7746
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.676786375416513e-06,
      "loss": 1.1358,
      "step": 7747
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.67530544242873e-06,
      "loss": 1.613,
      "step": 7748
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.67382450944095e-06,
      "loss": 1.227,
      "step": 7749
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.672343576453166e-06,
      "loss": 1.2382,
      "step": 7750
    },
    {
      "epoch": 1.71,
      "eval_loss": NaN,
      "eval_runtime": 4.3743,
      "eval_samples_per_second": 352.969,
      "eval_steps_per_second": 44.121,
      "step": 7750
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.670862643465385e-06,
      "loss": 1.302,
      "step": 7751
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.6693817104776e-06,
      "loss": 1.4807,
      "step": 7752
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.66790077748982e-06,
      "loss": 1.2634,
      "step": 7753
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.666419844502036e-06,
      "loss": 1.2857,
      "step": 7754
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.664938911514255e-06,
      "loss": 1.127,
      "step": 7755
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.663457978526472e-06,
      "loss": 1.4563,
      "step": 7756
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.66197704553869e-06,
      "loss": 1.2179,
      "step": 7757
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.660496112550907e-06,
      "loss": 1.343,
      "step": 7758
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.659015179563125e-06,
      "loss": 1.3248,
      "step": 7759
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.657534246575343e-06,
      "loss": 1.3143,
      "step": 7760
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.656053313587562e-06,
      "loss": 1.2516,
      "step": 7761
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.654572380599778e-06,
      "loss": 1.2409,
      "step": 7762
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.653091447611997e-06,
      "loss": 1.1042,
      "step": 7763
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.651610514624214e-06,
      "loss": 1.2072,
      "step": 7764
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.650129581636432e-06,
      "loss": 1.3775,
      "step": 7765
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.64864864864865e-06,
      "loss": 1.2418,
      "step": 7766
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.647167715660867e-06,
      "loss": 1.2792,
      "step": 7767
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.645686782673084e-06,
      "loss": 1.0991,
      "step": 7768
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.644205849685302e-06,
      "loss": 1.1997,
      "step": 7769
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.642724916697521e-06,
      "loss": 1.2428,
      "step": 7770
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.641243983709739e-06,
      "loss": 1.0719,
      "step": 7771
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.639763050721956e-06,
      "loss": 1.4058,
      "step": 7772
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.638282117734174e-06,
      "loss": 1.3473,
      "step": 7773
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.636801184746391e-06,
      "loss": 1.2762,
      "step": 7774
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.635320251758609e-06,
      "loss": 1.2073,
      "step": 7775
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.633839318770826e-06,
      "loss": 1.224,
      "step": 7776
    },
    {
      "epoch": 1.71,
      "learning_rate": 8.632358385783044e-06,
      "loss": 1.4165,
      "step": 7777
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.630877452795263e-06,
      "loss": 1.4524,
      "step": 7778
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.629396519807479e-06,
      "loss": 1.4075,
      "step": 7779
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.627915586819698e-06,
      "loss": 1.1596,
      "step": 7780
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.626434653831914e-06,
      "loss": 1.3043,
      "step": 7781
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.624953720844133e-06,
      "loss": 1.337,
      "step": 7782
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.62347278785635e-06,
      "loss": 1.2358,
      "step": 7783
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.621991854868568e-06,
      "loss": 1.3194,
      "step": 7784
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.620510921880786e-06,
      "loss": 1.2737,
      "step": 7785
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.619029988893003e-06,
      "loss": 1.4522,
      "step": 7786
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.61754905590522e-06,
      "loss": 1.3586,
      "step": 7787
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.61606812291744e-06,
      "loss": 1.499,
      "step": 7788
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.614587189929656e-06,
      "loss": 1.481,
      "step": 7789
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.613106256941875e-06,
      "loss": 1.3644,
      "step": 7790
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.61162532395409e-06,
      "loss": 1.3106,
      "step": 7791
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.61014439096631e-06,
      "loss": 1.4108,
      "step": 7792
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.608663457978527e-06,
      "loss": 1.3068,
      "step": 7793
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.607182524990745e-06,
      "loss": 1.1587,
      "step": 7794
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.605701592002962e-06,
      "loss": 1.3589,
      "step": 7795
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.60422065901518e-06,
      "loss": 1.0589,
      "step": 7796
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.602739726027397e-06,
      "loss": 1.4741,
      "step": 7797
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.601258793039615e-06,
      "loss": 1.2706,
      "step": 7798
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.599777860051834e-06,
      "loss": 1.2493,
      "step": 7799
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.598296927064052e-06,
      "loss": 1.404,
      "step": 7800
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.59681599407627e-06,
      "loss": 1.3647,
      "step": 7801
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.595335061088487e-06,
      "loss": 1.1212,
      "step": 7802
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.593854128100704e-06,
      "loss": 1.1388,
      "step": 7803
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.592373195112922e-06,
      "loss": 1.3169,
      "step": 7804
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.59089226212514e-06,
      "loss": 1.1622,
      "step": 7805
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.589411329137357e-06,
      "loss": 1.2947,
      "step": 7806
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.587930396149576e-06,
      "loss": 1.1869,
      "step": 7807
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.586449463161792e-06,
      "loss": 1.5841,
      "step": 7808
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.584968530174011e-06,
      "loss": 1.0848,
      "step": 7809
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.583487597186227e-06,
      "loss": 1.2315,
      "step": 7810
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.582006664198446e-06,
      "loss": 1.3268,
      "step": 7811
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.580525731210664e-06,
      "loss": 1.2962,
      "step": 7812
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.579044798222881e-06,
      "loss": 1.1227,
      "step": 7813
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.577563865235099e-06,
      "loss": 1.3893,
      "step": 7814
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.576082932247316e-06,
      "loss": 1.2941,
      "step": 7815
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.574601999259534e-06,
      "loss": 1.4199,
      "step": 7816
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.573121066271753e-06,
      "loss": 1.3029,
      "step": 7817
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.571640133283969e-06,
      "loss": 1.4691,
      "step": 7818
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.570159200296188e-06,
      "loss": 1.2633,
      "step": 7819
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.568678267308405e-06,
      "loss": 1.1751,
      "step": 7820
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.567197334320623e-06,
      "loss": 1.1926,
      "step": 7821
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.56571640133284e-06,
      "loss": 1.3689,
      "step": 7822
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.564235468345058e-06,
      "loss": 1.3255,
      "step": 7823
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.562754535357275e-06,
      "loss": 1.2254,
      "step": 7824
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.561273602369493e-06,
      "loss": 1.354,
      "step": 7825
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.55979266938171e-06,
      "loss": 1.2958,
      "step": 7826
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.55831173639393e-06,
      "loss": 1.6355,
      "step": 7827
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.556830803406147e-06,
      "loss": 1.3319,
      "step": 7828
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.555349870418365e-06,
      "loss": 1.3498,
      "step": 7829
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.553868937430582e-06,
      "loss": 1.452,
      "step": 7830
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.5523880044428e-06,
      "loss": 1.5568,
      "step": 7831
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.550907071455017e-06,
      "loss": 1.1576,
      "step": 7832
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.549426138467235e-06,
      "loss": 1.1386,
      "step": 7833
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.547945205479454e-06,
      "loss": 1.1661,
      "step": 7834
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.54646427249167e-06,
      "loss": 1.2795,
      "step": 7835
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.544983339503889e-06,
      "loss": 1.4421,
      "step": 7836
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.543502406516105e-06,
      "loss": 1.3896,
      "step": 7837
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.542021473528324e-06,
      "loss": 1.3318,
      "step": 7838
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.540540540540542e-06,
      "loss": 1.1393,
      "step": 7839
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.539059607552759e-06,
      "loss": 1.4555,
      "step": 7840
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.537578674564977e-06,
      "loss": 1.398,
      "step": 7841
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.536097741577194e-06,
      "loss": 1.248,
      "step": 7842
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.534616808589412e-06,
      "loss": 1.1819,
      "step": 7843
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.533135875601631e-06,
      "loss": 1.3654,
      "step": 7844
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.531654942613847e-06,
      "loss": 1.1866,
      "step": 7845
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.530174009626066e-06,
      "loss": 1.498,
      "step": 7846
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.528693076638282e-06,
      "loss": 1.247,
      "step": 7847
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.527212143650501e-06,
      "loss": 1.2138,
      "step": 7848
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.525731210662718e-06,
      "loss": 1.2253,
      "step": 7849
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.524250277674936e-06,
      "loss": 1.3175,
      "step": 7850
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.522769344687153e-06,
      "loss": 1.3972,
      "step": 7851
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.521288411699371e-06,
      "loss": 1.4183,
      "step": 7852
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.519807478711589e-06,
      "loss": 1.2667,
      "step": 7853
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.518326545723806e-06,
      "loss": 1.1026,
      "step": 7854
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.516845612736024e-06,
      "loss": 1.484,
      "step": 7855
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.515364679748243e-06,
      "loss": 1.2026,
      "step": 7856
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.51388374676046e-06,
      "loss": 1.5096,
      "step": 7857
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.512402813772678e-06,
      "loss": 1.2607,
      "step": 7858
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.510921880784895e-06,
      "loss": 1.3033,
      "step": 7859
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.509440947797113e-06,
      "loss": 1.7046,
      "step": 7860
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.50796001480933e-06,
      "loss": 1.1263,
      "step": 7861
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.506479081821548e-06,
      "loss": 1.1585,
      "step": 7862
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.504998148833767e-06,
      "loss": 1.3985,
      "step": 7863
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.503517215845983e-06,
      "loss": 1.4907,
      "step": 7864
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.502036282858202e-06,
      "loss": 1.3005,
      "step": 7865
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.500555349870418e-06,
      "loss": 1.3722,
      "step": 7866
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.499074416882637e-06,
      "loss": 1.2312,
      "step": 7867
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.497593483894855e-06,
      "loss": 1.5984,
      "step": 7868
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.496112550907072e-06,
      "loss": 1.4296,
      "step": 7869
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.49463161791929e-06,
      "loss": 1.0638,
      "step": 7870
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.493150684931507e-06,
      "loss": 1.2239,
      "step": 7871
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.491669751943725e-06,
      "loss": 1.4955,
      "step": 7872
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.490188818955944e-06,
      "loss": 1.0932,
      "step": 7873
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.48870788596816e-06,
      "loss": 1.1606,
      "step": 7874
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.487226952980379e-06,
      "loss": 1.1798,
      "step": 7875
    },
    {
      "epoch": 1.74,
      "eval_loss": NaN,
      "eval_runtime": 4.368,
      "eval_samples_per_second": 353.48,
      "eval_steps_per_second": 44.185,
      "step": 7875
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.485746019992595e-06,
      "loss": 1.3778,
      "step": 7876
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.484265087004814e-06,
      "loss": 1.4346,
      "step": 7877
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.482784154017032e-06,
      "loss": 1.1441,
      "step": 7878
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.481303221029249e-06,
      "loss": 1.1913,
      "step": 7879
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.479822288041467e-06,
      "loss": 1.2786,
      "step": 7880
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.478341355053684e-06,
      "loss": 1.1826,
      "step": 7881
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.476860422065902e-06,
      "loss": 1.3368,
      "step": 7882
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.475379489078119e-06,
      "loss": 1.0028,
      "step": 7883
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.473898556090338e-06,
      "loss": 1.1559,
      "step": 7884
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.472417623102556e-06,
      "loss": 1.2394,
      "step": 7885
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.470936690114773e-06,
      "loss": 1.0795,
      "step": 7886
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.469455757126991e-06,
      "loss": 1.2792,
      "step": 7887
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.467974824139208e-06,
      "loss": 1.0958,
      "step": 7888
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.466493891151426e-06,
      "loss": 1.4928,
      "step": 7889
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.465012958163643e-06,
      "loss": 1.4019,
      "step": 7890
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.463532025175861e-06,
      "loss": 1.4306,
      "step": 7891
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.46205109218808e-06,
      "loss": 1.1677,
      "step": 7892
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.460570159200296e-06,
      "loss": 1.1153,
      "step": 7893
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.459089226212515e-06,
      "loss": 1.3106,
      "step": 7894
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.457608293224733e-06,
      "loss": 1.3505,
      "step": 7895
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.45612736023695e-06,
      "loss": 1.2901,
      "step": 7896
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.454646427249168e-06,
      "loss": 1.1259,
      "step": 7897
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.453165494261385e-06,
      "loss": 1.3315,
      "step": 7898
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.451684561273603e-06,
      "loss": 1.4629,
      "step": 7899
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.450203628285822e-06,
      "loss": 1.2681,
      "step": 7900
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.448722695298038e-06,
      "loss": 1.1955,
      "step": 7901
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.447241762310257e-06,
      "loss": 1.2637,
      "step": 7902
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.445760829322473e-06,
      "loss": 1.3772,
      "step": 7903
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.444279896334692e-06,
      "loss": 1.1407,
      "step": 7904
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.44279896334691e-06,
      "loss": 1.3678,
      "step": 7905
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.441318030359127e-06,
      "loss": 1.4362,
      "step": 7906
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.439837097371345e-06,
      "loss": 1.3412,
      "step": 7907
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.438356164383562e-06,
      "loss": 1.2869,
      "step": 7908
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.43687523139578e-06,
      "loss": 1.2309,
      "step": 7909
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.435394298407997e-06,
      "loss": 1.4884,
      "step": 7910
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.433913365420215e-06,
      "loss": 1.2836,
      "step": 7911
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.432432432432434e-06,
      "loss": 1.34,
      "step": 7912
    },
    {
      "epoch": 1.74,
      "learning_rate": 8.430951499444651e-06,
      "loss": 1.1372,
      "step": 7913
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.429470566456869e-06,
      "loss": 1.0823,
      "step": 7914
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.427989633469086e-06,
      "loss": 1.3773,
      "step": 7915
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.426508700481304e-06,
      "loss": 1.3573,
      "step": 7916
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.425027767493521e-06,
      "loss": 1.136,
      "step": 7917
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.423546834505739e-06,
      "loss": 1.2553,
      "step": 7918
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.422065901517957e-06,
      "loss": 1.2763,
      "step": 7919
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.420584968530174e-06,
      "loss": 1.4556,
      "step": 7920
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.419104035542393e-06,
      "loss": 1.2014,
      "step": 7921
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.417623102554609e-06,
      "loss": 1.3887,
      "step": 7922
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.416142169566828e-06,
      "loss": 1.218,
      "step": 7923
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.414661236579046e-06,
      "loss": 1.3839,
      "step": 7924
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.413180303591263e-06,
      "loss": 1.2824,
      "step": 7925
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.41169937060348e-06,
      "loss": 1.5137,
      "step": 7926
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.410218437615698e-06,
      "loss": 1.3543,
      "step": 7927
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.408737504627916e-06,
      "loss": 1.2314,
      "step": 7928
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.407256571640135e-06,
      "loss": 1.3392,
      "step": 7929
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.405775638652351e-06,
      "loss": 1.5075,
      "step": 7930
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.40429470566457e-06,
      "loss": 1.1139,
      "step": 7931
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.402813772676786e-06,
      "loss": 1.1758,
      "step": 7932
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.401332839689005e-06,
      "loss": 1.2977,
      "step": 7933
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.399851906701223e-06,
      "loss": 1.4438,
      "step": 7934
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.39837097371344e-06,
      "loss": 1.2826,
      "step": 7935
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.396890040725658e-06,
      "loss": 1.3121,
      "step": 7936
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.395409107737875e-06,
      "loss": 1.1487,
      "step": 7937
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.393928174750093e-06,
      "loss": 1.2446,
      "step": 7938
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.39244724176231e-06,
      "loss": 1.2921,
      "step": 7939
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.390966308774528e-06,
      "loss": 1.2404,
      "step": 7940
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.389485375786747e-06,
      "loss": 1.3575,
      "step": 7941
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.388004442798964e-06,
      "loss": 1.2299,
      "step": 7942
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.386523509811182e-06,
      "loss": 1.224,
      "step": 7943
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.3850425768234e-06,
      "loss": 1.0796,
      "step": 7944
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.383561643835617e-06,
      "loss": 1.1276,
      "step": 7945
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.382080710847835e-06,
      "loss": 1.2759,
      "step": 7946
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.380599777860052e-06,
      "loss": 1.2711,
      "step": 7947
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.379118844872271e-06,
      "loss": 1.4285,
      "step": 7948
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.377637911884487e-06,
      "loss": 1.2939,
      "step": 7949
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.376156978896706e-06,
      "loss": 1.2095,
      "step": 7950
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.374676045908924e-06,
      "loss": 1.2437,
      "step": 7951
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.373195112921141e-06,
      "loss": 1.1359,
      "step": 7952
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.371714179933359e-06,
      "loss": 1.23,
      "step": 7953
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.370233246945576e-06,
      "loss": 1.2437,
      "step": 7954
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.368752313957794e-06,
      "loss": 1.1892,
      "step": 7955
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.367271380970011e-06,
      "loss": 1.1565,
      "step": 7956
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.365790447982229e-06,
      "loss": 1.256,
      "step": 7957
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.364309514994448e-06,
      "loss": 1.2205,
      "step": 7958
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.362828582006664e-06,
      "loss": 1.1124,
      "step": 7959
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.361347649018883e-06,
      "loss": 1.0993,
      "step": 7960
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.359866716031099e-06,
      "loss": 1.144,
      "step": 7961
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.358385783043318e-06,
      "loss": 1.6448,
      "step": 7962
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.356904850055536e-06,
      "loss": 1.1997,
      "step": 7963
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.355423917067753e-06,
      "loss": 1.0825,
      "step": 7964
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.35394298407997e-06,
      "loss": 1.234,
      "step": 7965
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.352462051092188e-06,
      "loss": 1.1688,
      "step": 7966
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.350981118104406e-06,
      "loss": 1.2779,
      "step": 7967
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.349500185116625e-06,
      "loss": 1.1637,
      "step": 7968
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.348019252128843e-06,
      "loss": 1.3819,
      "step": 7969
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.34653831914106e-06,
      "loss": 1.3978,
      "step": 7970
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.345057386153278e-06,
      "loss": 1.4689,
      "step": 7971
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.343576453165495e-06,
      "loss": 1.1912,
      "step": 7972
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.342095520177713e-06,
      "loss": 1.2807,
      "step": 7973
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.34061458718993e-06,
      "loss": 1.1823,
      "step": 7974
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.339133654202148e-06,
      "loss": 1.3181,
      "step": 7975
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.337652721214365e-06,
      "loss": 1.6177,
      "step": 7976
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.336171788226584e-06,
      "loss": 1.1744,
      "step": 7977
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.3346908552388e-06,
      "loss": 1.164,
      "step": 7978
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.33320992225102e-06,
      "loss": 1.3439,
      "step": 7979
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.331728989263237e-06,
      "loss": 1.1415,
      "step": 7980
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.330248056275454e-06,
      "loss": 1.1707,
      "step": 7981
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.328767123287672e-06,
      "loss": 1.248,
      "step": 7982
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.32728619029989e-06,
      "loss": 1.4735,
      "step": 7983
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.325805257312107e-06,
      "loss": 1.4182,
      "step": 7984
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.324324324324326e-06,
      "loss": 1.0814,
      "step": 7985
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.322843391336542e-06,
      "loss": 1.3452,
      "step": 7986
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.321362458348761e-06,
      "loss": 1.297,
      "step": 7987
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.319881525360977e-06,
      "loss": 1.1784,
      "step": 7988
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.318400592373196e-06,
      "loss": 1.1667,
      "step": 7989
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.316919659385414e-06,
      "loss": 1.291,
      "step": 7990
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.315438726397631e-06,
      "loss": 1.2252,
      "step": 7991
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.313957793409849e-06,
      "loss": 1.1782,
      "step": 7992
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.312476860422066e-06,
      "loss": 1.3739,
      "step": 7993
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.310995927434284e-06,
      "loss": 1.284,
      "step": 7994
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.309514994446501e-06,
      "loss": 1.1011,
      "step": 7995
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.308034061458719e-06,
      "loss": 1.3182,
      "step": 7996
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.306553128470938e-06,
      "loss": 1.1415,
      "step": 7997
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.305072195483156e-06,
      "loss": 1.2694,
      "step": 7998
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.303591262495373e-06,
      "loss": 1.2716,
      "step": 7999
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.30211032950759e-06,
      "loss": 1.4486,
      "step": 8000
    },
    {
      "epoch": 1.76,
      "eval_loss": NaN,
      "eval_runtime": 4.3774,
      "eval_samples_per_second": 352.724,
      "eval_steps_per_second": 44.09,
      "step": 8000
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.300629396519808e-06,
      "loss": 1.1774,
      "step": 8001
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.299148463532026e-06,
      "loss": 1.3504,
      "step": 8002
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.297667530544243e-06,
      "loss": 1.1855,
      "step": 8003
    },
    {
      "epoch": 1.76,
      "learning_rate": 8.29618659755646e-06,
      "loss": 1.3422,
      "step": 8004
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.294705664568678e-06,
      "loss": 1.4287,
      "step": 8005
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.293224731580897e-06,
      "loss": 1.5243,
      "step": 8006
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.291743798593115e-06,
      "loss": 1.1708,
      "step": 8007
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.290262865605332e-06,
      "loss": 1.0773,
      "step": 8008
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.28878193261755e-06,
      "loss": 1.1409,
      "step": 8009
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.287300999629767e-06,
      "loss": 1.1598,
      "step": 8010
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.285820066641985e-06,
      "loss": 1.4654,
      "step": 8011
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.284339133654203e-06,
      "loss": 1.2783,
      "step": 8012
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.28285820066642e-06,
      "loss": 1.3306,
      "step": 8013
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.28137726767864e-06,
      "loss": 1.2944,
      "step": 8014
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.279896334690855e-06,
      "loss": 1.1644,
      "step": 8015
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.278415401703074e-06,
      "loss": 1.1404,
      "step": 8016
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.27693446871529e-06,
      "loss": 1.5427,
      "step": 8017
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.27545353572751e-06,
      "loss": 1.4164,
      "step": 8018
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.273972602739727e-06,
      "loss": 1.2625,
      "step": 8019
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.272491669751944e-06,
      "loss": 1.4278,
      "step": 8020
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.271010736764162e-06,
      "loss": 1.4121,
      "step": 8021
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.26952980377638e-06,
      "loss": 1.2731,
      "step": 8022
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.268048870788597e-06,
      "loss": 1.2789,
      "step": 8023
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.266567937800816e-06,
      "loss": 1.3647,
      "step": 8024
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.265087004813032e-06,
      "loss": 1.2862,
      "step": 8025
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.263606071825251e-06,
      "loss": 1.1524,
      "step": 8026
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.262125138837469e-06,
      "loss": 1.1096,
      "step": 8027
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.260644205849686e-06,
      "loss": 1.1175,
      "step": 8028
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.259163272861904e-06,
      "loss": 1.3845,
      "step": 8029
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.257682339874121e-06,
      "loss": 1.5621,
      "step": 8030
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.256201406886339e-06,
      "loss": 1.2958,
      "step": 8031
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.254720473898556e-06,
      "loss": 1.288,
      "step": 8032
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.253239540910775e-06,
      "loss": 1.2033,
      "step": 8033
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.251758607922991e-06,
      "loss": 1.3011,
      "step": 8034
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.25027767493521e-06,
      "loss": 1.231,
      "step": 8035
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.248796741947428e-06,
      "loss": 1.3127,
      "step": 8036
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.247315808959646e-06,
      "loss": 1.3413,
      "step": 8037
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.245834875971863e-06,
      "loss": 1.1786,
      "step": 8038
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.24435394298408e-06,
      "loss": 1.2772,
      "step": 8039
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.242873009996298e-06,
      "loss": 1.3589,
      "step": 8040
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.241392077008517e-06,
      "loss": 1.4521,
      "step": 8041
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.239911144020733e-06,
      "loss": 1.2399,
      "step": 8042
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.238430211032952e-06,
      "loss": 1.246,
      "step": 8043
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.236949278045168e-06,
      "loss": 1.1661,
      "step": 8044
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.235468345057387e-06,
      "loss": 1.4102,
      "step": 8045
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.233987412069605e-06,
      "loss": 1.1169,
      "step": 8046
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.232506479081822e-06,
      "loss": 1.4633,
      "step": 8047
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.23102554609404e-06,
      "loss": 1.1689,
      "step": 8048
    },
    {
      "epoch": 1.77,
      "learning_rate": 8.229544613106257e-06,
      "loss": 1.0209,
      "step": 8049
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.228063680118475e-06,
      "loss": 1.0087,
      "step": 8050
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.226582747130692e-06,
      "loss": 1.3007,
      "step": 8051
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.22510181414291e-06,
      "loss": 1.2847,
      "step": 8052
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.22362088115513e-06,
      "loss": 1.3734,
      "step": 8053
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.222139948167345e-06,
      "loss": 1.4314,
      "step": 8054
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.220659015179564e-06,
      "loss": 1.149,
      "step": 8055
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.219178082191782e-06,
      "loss": 1.1868,
      "step": 8056
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.217697149204e-06,
      "loss": 1.2974,
      "step": 8057
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.216216216216217e-06,
      "loss": 1.3964,
      "step": 8058
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.214735283228434e-06,
      "loss": 1.3563,
      "step": 8059
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.213254350240652e-06,
      "loss": 1.2285,
      "step": 8060
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.21177341725287e-06,
      "loss": 1.1315,
      "step": 8061
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.210292484265089e-06,
      "loss": 1.3239,
      "step": 8062
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.208811551277306e-06,
      "loss": 1.072,
      "step": 8063
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.207330618289524e-06,
      "loss": 1.3522,
      "step": 8064
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.205849685301741e-06,
      "loss": 1.264,
      "step": 8065
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.204368752313959e-06,
      "loss": 1.3699,
      "step": 8066
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.202887819326176e-06,
      "loss": 1.1961,
      "step": 8067
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.201406886338394e-06,
      "loss": 1.4035,
      "step": 8068
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.199925953350611e-06,
      "loss": 1.2066,
      "step": 8069
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.19844502036283e-06,
      "loss": 1.3356,
      "step": 8070
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.196964087375046e-06,
      "loss": 1.1532,
      "step": 8071
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.195483154387265e-06,
      "loss": 0.9977,
      "step": 8072
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.194002221399481e-06,
      "loss": 1.1666,
      "step": 8073
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.1925212884117e-06,
      "loss": 1.4688,
      "step": 8074
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.191040355423918e-06,
      "loss": 1.2357,
      "step": 8075
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.189559422436135e-06,
      "loss": 1.2348,
      "step": 8076
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.188078489448353e-06,
      "loss": 1.0718,
      "step": 8077
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.18659755646057e-06,
      "loss": 1.3973,
      "step": 8078
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.185116623472788e-06,
      "loss": 1.09,
      "step": 8079
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.183635690485007e-06,
      "loss": 1.4096,
      "step": 8080
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.182154757497223e-06,
      "loss": 1.4629,
      "step": 8081
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.180673824509442e-06,
      "loss": 1.2344,
      "step": 8082
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.17919289152166e-06,
      "loss": 1.1119,
      "step": 8083
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.177711958533877e-06,
      "loss": 1.0676,
      "step": 8084
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.176231025546095e-06,
      "loss": 1.2884,
      "step": 8085
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.174750092558312e-06,
      "loss": 1.1597,
      "step": 8086
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.17326915957053e-06,
      "loss": 1.4422,
      "step": 8087
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.171788226582747e-06,
      "loss": 1.0419,
      "step": 8088
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.170307293594965e-06,
      "loss": 1.0965,
      "step": 8089
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.168826360607182e-06,
      "loss": 1.3368,
      "step": 8090
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.167345427619402e-06,
      "loss": 1.3726,
      "step": 8091
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.165864494631619e-06,
      "loss": 1.3373,
      "step": 8092
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.164383561643837e-06,
      "loss": 1.2769,
      "step": 8093
    },
    {
      "epoch": 1.78,
      "learning_rate": 8.162902628656054e-06,
      "loss": 1.1833,
      "step": 8094
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.161421695668272e-06,
      "loss": 1.3006,
      "step": 8095
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.15994076268049e-06,
      "loss": 1.4757,
      "step": 8096
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.158459829692708e-06,
      "loss": 1.3691,
      "step": 8097
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.156978896704924e-06,
      "loss": 1.2349,
      "step": 8098
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.155497963717143e-06,
      "loss": 1.2459,
      "step": 8099
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.15401703072936e-06,
      "loss": 1.5527,
      "step": 8100
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.152536097741578e-06,
      "loss": 1.3511,
      "step": 8101
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.151055164753796e-06,
      "loss": 1.3295,
      "step": 8102
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.149574231766013e-06,
      "loss": 1.1104,
      "step": 8103
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.148093298778231e-06,
      "loss": 1.22,
      "step": 8104
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.146612365790449e-06,
      "loss": 1.1974,
      "step": 8105
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.145131432802666e-06,
      "loss": 1.3666,
      "step": 8106
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.143650499814884e-06,
      "loss": 1.4837,
      "step": 8107
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.142169566827101e-06,
      "loss": 1.3811,
      "step": 8108
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.14068863383932e-06,
      "loss": 1.4187,
      "step": 8109
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.139207700851536e-06,
      "loss": 1.154,
      "step": 8110
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.137726767863755e-06,
      "loss": 1.2076,
      "step": 8111
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.136245834875973e-06,
      "loss": 1.0305,
      "step": 8112
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.13476490188819e-06,
      "loss": 1.2952,
      "step": 8113
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.133283968900408e-06,
      "loss": 1.3979,
      "step": 8114
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.131803035912625e-06,
      "loss": 1.1384,
      "step": 8115
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.130322102924843e-06,
      "loss": 1.238,
      "step": 8116
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.12884116993706e-06,
      "loss": 1.5124,
      "step": 8117
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.127360236949278e-06,
      "loss": 1.3031,
      "step": 8118
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.125879303961497e-06,
      "loss": 1.0648,
      "step": 8119
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.124398370973715e-06,
      "loss": 1.2008,
      "step": 8120
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.122917437985932e-06,
      "loss": 1.456,
      "step": 8121
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.12143650499815e-06,
      "loss": 1.4327,
      "step": 8122
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.119955572010367e-06,
      "loss": 1.3702,
      "step": 8123
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.118474639022585e-06,
      "loss": 1.1147,
      "step": 8124
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.116993706034802e-06,
      "loss": 1.4267,
      "step": 8125
    },
    {
      "epoch": 1.79,
      "eval_loss": NaN,
      "eval_runtime": 4.3747,
      "eval_samples_per_second": 352.938,
      "eval_steps_per_second": 44.117,
      "step": 8125
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.115512773047021e-06,
      "loss": 1.4195,
      "step": 8126
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.114031840059237e-06,
      "loss": 1.3053,
      "step": 8127
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.112550907071456e-06,
      "loss": 1.4191,
      "step": 8128
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.111069974083672e-06,
      "loss": 1.4907,
      "step": 8129
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.109589041095892e-06,
      "loss": 1.2992,
      "step": 8130
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.108108108108109e-06,
      "loss": 1.3957,
      "step": 8131
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.106627175120327e-06,
      "loss": 1.2371,
      "step": 8132
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.105146242132544e-06,
      "loss": 1.1118,
      "step": 8133
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.103665309144762e-06,
      "loss": 1.1611,
      "step": 8134
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.102184376156979e-06,
      "loss": 1.0956,
      "step": 8135
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.100703443169198e-06,
      "loss": 1.3517,
      "step": 8136
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.099222510181414e-06,
      "loss": 1.2713,
      "step": 8137
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.097741577193633e-06,
      "loss": 1.4783,
      "step": 8138
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.09626064420585e-06,
      "loss": 1.3038,
      "step": 8139
    },
    {
      "epoch": 1.79,
      "learning_rate": 8.094779711218068e-06,
      "loss": 1.2667,
      "step": 8140
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.093298778230286e-06,
      "loss": 1.4031,
      "step": 8141
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.091817845242503e-06,
      "loss": 1.2988,
      "step": 8142
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.090336912254721e-06,
      "loss": 1.1938,
      "step": 8143
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.088855979266938e-06,
      "loss": 1.2447,
      "step": 8144
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.087375046279156e-06,
      "loss": 1.3126,
      "step": 8145
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.085894113291373e-06,
      "loss": 1.1649,
      "step": 8146
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.084413180303593e-06,
      "loss": 1.5303,
      "step": 8147
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.08293224731581e-06,
      "loss": 1.2737,
      "step": 8148
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.081451314328028e-06,
      "loss": 1.3231,
      "step": 8149
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.079970381340245e-06,
      "loss": 1.3184,
      "step": 8150
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.078489448352463e-06,
      "loss": 1.2151,
      "step": 8151
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.07700851536468e-06,
      "loss": 1.0854,
      "step": 8152
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.075527582376898e-06,
      "loss": 1.2996,
      "step": 8153
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.074046649389115e-06,
      "loss": 1.4781,
      "step": 8154
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.072565716401335e-06,
      "loss": 1.4331,
      "step": 8155
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.07108478341355e-06,
      "loss": 1.3883,
      "step": 8156
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.06960385042577e-06,
      "loss": 1.3612,
      "step": 8157
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.068122917437987e-06,
      "loss": 1.2887,
      "step": 8158
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.066641984450205e-06,
      "loss": 1.2215,
      "step": 8159
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.065161051462422e-06,
      "loss": 1.1316,
      "step": 8160
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.06368011847464e-06,
      "loss": 1.392,
      "step": 8161
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.062199185486857e-06,
      "loss": 1.3373,
      "step": 8162
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.060718252499075e-06,
      "loss": 1.0472,
      "step": 8163
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.059237319511292e-06,
      "loss": 1.5471,
      "step": 8164
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.057756386523511e-06,
      "loss": 1.1978,
      "step": 8165
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.056275453535727e-06,
      "loss": 1.1669,
      "step": 8166
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.054794520547946e-06,
      "loss": 1.3364,
      "step": 8167
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.053313587560164e-06,
      "loss": 1.3066,
      "step": 8168
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.051832654572381e-06,
      "loss": 1.3778,
      "step": 8169
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.050351721584599e-06,
      "loss": 1.4009,
      "step": 8170
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.048870788596816e-06,
      "loss": 1.2802,
      "step": 8171
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.047389855609034e-06,
      "loss": 1.0619,
      "step": 8172
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.045908922621252e-06,
      "loss": 1.2314,
      "step": 8173
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.044427989633469e-06,
      "loss": 1.3014,
      "step": 8174
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.042947056645688e-06,
      "loss": 1.135,
      "step": 8175
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.041466123657906e-06,
      "loss": 1.2696,
      "step": 8176
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.039985190670123e-06,
      "loss": 1.4197,
      "step": 8177
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.03850425768234e-06,
      "loss": 1.1614,
      "step": 8178
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.037023324694558e-06,
      "loss": 1.1227,
      "step": 8179
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.035542391706776e-06,
      "loss": 1.1755,
      "step": 8180
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.034061458718993e-06,
      "loss": 1.2351,
      "step": 8181
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.032580525731211e-06,
      "loss": 1.5097,
      "step": 8182
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.031099592743428e-06,
      "loss": 1.182,
      "step": 8183
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.029618659755648e-06,
      "loss": 1.3942,
      "step": 8184
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.028137726767863e-06,
      "loss": 1.1409,
      "step": 8185
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.026656793780083e-06,
      "loss": 1.0392,
      "step": 8186
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.0251758607923e-06,
      "loss": 1.2009,
      "step": 8187
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.023694927804518e-06,
      "loss": 1.3963,
      "step": 8188
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.022213994816735e-06,
      "loss": 1.2961,
      "step": 8189
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.020733061828953e-06,
      "loss": 1.2464,
      "step": 8190
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.01925212884117e-06,
      "loss": 1.1552,
      "step": 8191
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.01777119585339e-06,
      "loss": 1.3803,
      "step": 8192
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.016290262865605e-06,
      "loss": 1.1672,
      "step": 8193
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.014809329877824e-06,
      "loss": 1.2393,
      "step": 8194
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.01332839689004e-06,
      "loss": 1.2939,
      "step": 8195
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.01184746390226e-06,
      "loss": 1.3727,
      "step": 8196
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.010366530914477e-06,
      "loss": 1.3821,
      "step": 8197
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.008885597926695e-06,
      "loss": 1.4434,
      "step": 8198
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.007404664938912e-06,
      "loss": 1.1906,
      "step": 8199
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.00592373195113e-06,
      "loss": 1.0977,
      "step": 8200
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.004442798963347e-06,
      "loss": 0.9961,
      "step": 8201
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.002961865975565e-06,
      "loss": 1.3421,
      "step": 8202
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.001480932987782e-06,
      "loss": 1.3244,
      "step": 8203
    },
    {
      "epoch": 1.81,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.2709,
      "step": 8204
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.998519067012219e-06,
      "loss": 1.325,
      "step": 8205
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.997038134024436e-06,
      "loss": 1.3051,
      "step": 8206
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.995557201036654e-06,
      "loss": 1.1924,
      "step": 8207
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.994076268048871e-06,
      "loss": 1.3965,
      "step": 8208
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.992595335061089e-06,
      "loss": 1.2731,
      "step": 8209
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.991114402073306e-06,
      "loss": 1.248,
      "step": 8210
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.989633469085526e-06,
      "loss": 1.218,
      "step": 8211
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.988152536097741e-06,
      "loss": 1.3131,
      "step": 8212
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.98667160310996e-06,
      "loss": 1.3161,
      "step": 8213
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.985190670122177e-06,
      "loss": 1.4978,
      "step": 8214
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.983709737134396e-06,
      "loss": 1.64,
      "step": 8215
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.982228804146613e-06,
      "loss": 1.261,
      "step": 8216
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.98074787115883e-06,
      "loss": 1.3095,
      "step": 8217
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.979266938171048e-06,
      "loss": 1.3856,
      "step": 8218
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.977786005183266e-06,
      "loss": 1.282,
      "step": 8219
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.976305072195483e-06,
      "loss": 1.1639,
      "step": 8220
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.974824139207703e-06,
      "loss": 1.3088,
      "step": 8221
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.973343206219918e-06,
      "loss": 1.3383,
      "step": 8222
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.971862273232138e-06,
      "loss": 1.308,
      "step": 8223
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.970381340244353e-06,
      "loss": 1.3138,
      "step": 8224
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.968900407256573e-06,
      "loss": 1.3497,
      "step": 8225
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.96741947426879e-06,
      "loss": 1.14,
      "step": 8226
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.965938541281008e-06,
      "loss": 1.2076,
      "step": 8227
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.964457608293225e-06,
      "loss": 1.3664,
      "step": 8228
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.962976675305443e-06,
      "loss": 1.2976,
      "step": 8229
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.96149574231766e-06,
      "loss": 1.3452,
      "step": 8230
    },
    {
      "epoch": 1.81,
      "learning_rate": 7.96001480932988e-06,
      "loss": 1.2952,
      "step": 8231
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.958533876342097e-06,
      "loss": 1.2,
      "step": 8232
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.957052943354314e-06,
      "loss": 1.3022,
      "step": 8233
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.955572010366532e-06,
      "loss": 1.4582,
      "step": 8234
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.95409107737875e-06,
      "loss": 1.345,
      "step": 8235
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.952610144390967e-06,
      "loss": 1.2854,
      "step": 8236
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.951129211403184e-06,
      "loss": 1.3985,
      "step": 8237
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.949648278415402e-06,
      "loss": 1.2733,
      "step": 8238
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.94816734542762e-06,
      "loss": 1.2634,
      "step": 8239
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.946686412439839e-06,
      "loss": 1.3429,
      "step": 8240
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.945205479452055e-06,
      "loss": 1.2993,
      "step": 8241
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.943724546464274e-06,
      "loss": 1.1104,
      "step": 8242
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.942243613476491e-06,
      "loss": 1.3039,
      "step": 8243
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.940762680488709e-06,
      "loss": 1.3321,
      "step": 8244
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.939281747500926e-06,
      "loss": 1.2426,
      "step": 8245
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.937800814513144e-06,
      "loss": 1.2439,
      "step": 8246
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.936319881525361e-06,
      "loss": 1.2207,
      "step": 8247
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.93483894853758e-06,
      "loss": 1.4359,
      "step": 8248
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.933358015549796e-06,
      "loss": 1.2368,
      "step": 8249
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.931877082562016e-06,
      "loss": 1.2768,
      "step": 8250
    },
    {
      "epoch": 1.82,
      "eval_loss": NaN,
      "eval_runtime": 4.3762,
      "eval_samples_per_second": 352.818,
      "eval_steps_per_second": 44.102,
      "step": 8250
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.930396149574231e-06,
      "loss": 1.315,
      "step": 8251
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.92891521658645e-06,
      "loss": 1.1593,
      "step": 8252
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.927434283598668e-06,
      "loss": 1.2854,
      "step": 8253
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.925953350610886e-06,
      "loss": 1.512,
      "step": 8254
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.924472417623103e-06,
      "loss": 1.0862,
      "step": 8255
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.92299148463532e-06,
      "loss": 1.3712,
      "step": 8256
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.921510551647538e-06,
      "loss": 1.4585,
      "step": 8257
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.920029618659756e-06,
      "loss": 1.1744,
      "step": 8258
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.918548685671973e-06,
      "loss": 1.1546,
      "step": 8259
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.917067752684192e-06,
      "loss": 1.2496,
      "step": 8260
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.91558681969641e-06,
      "loss": 1.4312,
      "step": 8261
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.914105886708627e-06,
      "loss": 1.2882,
      "step": 8262
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.912624953720845e-06,
      "loss": 1.4733,
      "step": 8263
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.911144020733063e-06,
      "loss": 1.3858,
      "step": 8264
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.90966308774528e-06,
      "loss": 1.2164,
      "step": 8265
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.908182154757498e-06,
      "loss": 1.3738,
      "step": 8266
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.906701221769715e-06,
      "loss": 1.2938,
      "step": 8267
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.905220288781933e-06,
      "loss": 1.226,
      "step": 8268
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.903739355794152e-06,
      "loss": 1.0804,
      "step": 8269
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.902258422806368e-06,
      "loss": 1.2498,
      "step": 8270
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.900777489818587e-06,
      "loss": 1.4224,
      "step": 8271
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.899296556830804e-06,
      "loss": 1.5036,
      "step": 8272
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.897815623843022e-06,
      "loss": 1.3473,
      "step": 8273
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.89633469085524e-06,
      "loss": 1.406,
      "step": 8274
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.894853757867457e-06,
      "loss": 1.2846,
      "step": 8275
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.893372824879674e-06,
      "loss": 1.2234,
      "step": 8276
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.891891891891894e-06,
      "loss": 1.339,
      "step": 8277
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.89041095890411e-06,
      "loss": 1.3734,
      "step": 8278
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.888930025916329e-06,
      "loss": 1.3016,
      "step": 8279
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.887449092928544e-06,
      "loss": 1.3225,
      "step": 8280
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.885968159940764e-06,
      "loss": 1.3113,
      "step": 8281
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.884487226952981e-06,
      "loss": 1.3222,
      "step": 8282
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.883006293965199e-06,
      "loss": 1.3374,
      "step": 8283
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.881525360977416e-06,
      "loss": 1.2356,
      "step": 8284
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.880044427989634e-06,
      "loss": 1.1579,
      "step": 8285
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.878563495001851e-06,
      "loss": 1.3531,
      "step": 8286
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.877082562014069e-06,
      "loss": 1.3669,
      "step": 8287
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.875601629026286e-06,
      "loss": 1.5297,
      "step": 8288
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.874120696038506e-06,
      "loss": 1.3687,
      "step": 8289
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.872639763050723e-06,
      "loss": 1.3647,
      "step": 8290
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.87115883006294e-06,
      "loss": 1.3657,
      "step": 8291
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.869677897075158e-06,
      "loss": 1.3015,
      "step": 8292
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.868196964087376e-06,
      "loss": 1.4937,
      "step": 8293
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.866716031099593e-06,
      "loss": 1.1648,
      "step": 8294
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.86523509811181e-06,
      "loss": 1.1471,
      "step": 8295
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.86375416512403e-06,
      "loss": 1.4643,
      "step": 8296
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.862273232136246e-06,
      "loss": 1.3087,
      "step": 8297
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.860792299148465e-06,
      "loss": 1.1996,
      "step": 8298
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.859311366160682e-06,
      "loss": 1.6684,
      "step": 8299
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.8578304331729e-06,
      "loss": 1.3149,
      "step": 8300
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.856349500185117e-06,
      "loss": 1.2583,
      "step": 8301
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.854868567197335e-06,
      "loss": 1.1441,
      "step": 8302
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.853387634209552e-06,
      "loss": 1.2888,
      "step": 8303
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.851906701221772e-06,
      "loss": 1.1225,
      "step": 8304
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.850425768233987e-06,
      "loss": 1.4336,
      "step": 8305
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.848944835246207e-06,
      "loss": 1.3357,
      "step": 8306
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.847463902258423e-06,
      "loss": 1.328,
      "step": 8307
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.845982969270642e-06,
      "loss": 1.2957,
      "step": 8308
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.844502036282858e-06,
      "loss": 1.1099,
      "step": 8309
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.843021103295077e-06,
      "loss": 1.2739,
      "step": 8310
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.841540170307294e-06,
      "loss": 1.3674,
      "step": 8311
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.840059237319512e-06,
      "loss": 1.4272,
      "step": 8312
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.83857830433173e-06,
      "loss": 1.3705,
      "step": 8313
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.837097371343947e-06,
      "loss": 1.2477,
      "step": 8314
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.835616438356164e-06,
      "loss": 1.2419,
      "step": 8315
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.834135505368384e-06,
      "loss": 1.4958,
      "step": 8316
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.832654572380601e-06,
      "loss": 1.3537,
      "step": 8317
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.831173639392819e-06,
      "loss": 1.3015,
      "step": 8318
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.829692706405036e-06,
      "loss": 1.3053,
      "step": 8319
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.828211773417254e-06,
      "loss": 1.3496,
      "step": 8320
    },
    {
      "epoch": 1.83,
      "learning_rate": 7.826730840429471e-06,
      "loss": 1.1445,
      "step": 8321
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.825249907441689e-06,
      "loss": 1.4073,
      "step": 8322
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.823768974453906e-06,
      "loss": 1.2739,
      "step": 8323
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.822288041466124e-06,
      "loss": 1.39,
      "step": 8324
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.820807108478343e-06,
      "loss": 1.3885,
      "step": 8325
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.819326175490559e-06,
      "loss": 1.5241,
      "step": 8326
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.817845242502778e-06,
      "loss": 1.2881,
      "step": 8327
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.816364309514995e-06,
      "loss": 1.2202,
      "step": 8328
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.814883376527213e-06,
      "loss": 1.3564,
      "step": 8329
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.81340244353943e-06,
      "loss": 1.3028,
      "step": 8330
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.811921510551648e-06,
      "loss": 1.3472,
      "step": 8331
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.810440577563866e-06,
      "loss": 1.4333,
      "step": 8332
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.808959644576085e-06,
      "loss": 1.3511,
      "step": 8333
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.8074787115883e-06,
      "loss": 1.2832,
      "step": 8334
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.80599777860052e-06,
      "loss": 1.1238,
      "step": 8335
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.804516845612736e-06,
      "loss": 1.1165,
      "step": 8336
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.803035912624955e-06,
      "loss": 1.3255,
      "step": 8337
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.801554979637172e-06,
      "loss": 1.1204,
      "step": 8338
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.80007404664939e-06,
      "loss": 1.1951,
      "step": 8339
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.798593113661607e-06,
      "loss": 1.1156,
      "step": 8340
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.797112180673825e-06,
      "loss": 1.3772,
      "step": 8341
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.795631247686042e-06,
      "loss": 1.2943,
      "step": 8342
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.79415031469826e-06,
      "loss": 1.2321,
      "step": 8343
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.792669381710477e-06,
      "loss": 1.4328,
      "step": 8344
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.791188448722697e-06,
      "loss": 1.3397,
      "step": 8345
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.789707515734914e-06,
      "loss": 0.9242,
      "step": 8346
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.788226582747132e-06,
      "loss": 1.4473,
      "step": 8347
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.78674564975935e-06,
      "loss": 1.3307,
      "step": 8348
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.785264716771567e-06,
      "loss": 1.3077,
      "step": 8349
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.783783783783784e-06,
      "loss": 1.1068,
      "step": 8350
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.782302850796002e-06,
      "loss": 1.2118,
      "step": 8351
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.78082191780822e-06,
      "loss": 1.2023,
      "step": 8352
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.779340984820437e-06,
      "loss": 1.2535,
      "step": 8353
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.777860051832656e-06,
      "loss": 1.2851,
      "step": 8354
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.776379118844873e-06,
      "loss": 1.5288,
      "step": 8355
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.774898185857091e-06,
      "loss": 1.4569,
      "step": 8356
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.773417252869309e-06,
      "loss": 1.1521,
      "step": 8357
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.771936319881526e-06,
      "loss": 1.0913,
      "step": 8358
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.770455386893744e-06,
      "loss": 1.2515,
      "step": 8359
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.768974453905963e-06,
      "loss": 1.2497,
      "step": 8360
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.767493520918179e-06,
      "loss": 1.2709,
      "step": 8361
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.766012587930398e-06,
      "loss": 1.0712,
      "step": 8362
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.764531654942614e-06,
      "loss": 1.2836,
      "step": 8363
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.763050721954833e-06,
      "loss": 1.1778,
      "step": 8364
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.761569788967049e-06,
      "loss": 1.3678,
      "step": 8365
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.760088855979268e-06,
      "loss": 1.3696,
      "step": 8366
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.758607922991485e-06,
      "loss": 1.3308,
      "step": 8367
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.757126990003703e-06,
      "loss": 1.186,
      "step": 8368
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.75564605701592e-06,
      "loss": 1.3805,
      "step": 8369
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.754165124028138e-06,
      "loss": 1.3203,
      "step": 8370
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.752684191040355e-06,
      "loss": 1.2452,
      "step": 8371
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.751203258052575e-06,
      "loss": 1.3829,
      "step": 8372
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.74972232506479e-06,
      "loss": 1.2762,
      "step": 8373
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.74824139207701e-06,
      "loss": 1.3139,
      "step": 8374
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.746760459089227e-06,
      "loss": 1.3592,
      "step": 8375
    },
    {
      "epoch": 1.85,
      "eval_loss": NaN,
      "eval_runtime": 4.3688,
      "eval_samples_per_second": 353.417,
      "eval_steps_per_second": 44.177,
      "step": 8375
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.745279526101445e-06,
      "loss": 1.3742,
      "step": 8376
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.743798593113662e-06,
      "loss": 1.2676,
      "step": 8377
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.74231766012588e-06,
      "loss": 1.3518,
      "step": 8378
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.740836727138097e-06,
      "loss": 1.3905,
      "step": 8379
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.739355794150315e-06,
      "loss": 1.2797,
      "step": 8380
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.737874861162534e-06,
      "loss": 1.275,
      "step": 8381
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.73639392817475e-06,
      "loss": 1.414,
      "step": 8382
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.734912995186969e-06,
      "loss": 1.3368,
      "step": 8383
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.733432062199187e-06,
      "loss": 1.2537,
      "step": 8384
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.731951129211404e-06,
      "loss": 1.409,
      "step": 8385
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.730470196223622e-06,
      "loss": 1.3664,
      "step": 8386
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.728989263235839e-06,
      "loss": 1.1348,
      "step": 8387
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.727508330248057e-06,
      "loss": 1.2247,
      "step": 8388
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.726027397260276e-06,
      "loss": 1.2105,
      "step": 8389
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.724546464272492e-06,
      "loss": 1.3927,
      "step": 8390
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.723065531284711e-06,
      "loss": 1.1926,
      "step": 8391
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.721584598296927e-06,
      "loss": 1.3091,
      "step": 8392
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.720103665309146e-06,
      "loss": 1.2187,
      "step": 8393
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.718622732321363e-06,
      "loss": 1.316,
      "step": 8394
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.717141799333581e-06,
      "loss": 1.2994,
      "step": 8395
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.715660866345798e-06,
      "loss": 1.2942,
      "step": 8396
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.714179933358016e-06,
      "loss": 1.3708,
      "step": 8397
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.712699000370233e-06,
      "loss": 1.2665,
      "step": 8398
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.711218067382451e-06,
      "loss": 1.2796,
      "step": 8399
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.709737134394669e-06,
      "loss": 1.3524,
      "step": 8400
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.708256201406888e-06,
      "loss": 1.3706,
      "step": 8401
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.706775268419104e-06,
      "loss": 1.1713,
      "step": 8402
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.705294335431323e-06,
      "loss": 1.314,
      "step": 8403
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.70381340244354e-06,
      "loss": 1.1799,
      "step": 8404
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.702332469455758e-06,
      "loss": 1.2101,
      "step": 8405
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.700851536467975e-06,
      "loss": 1.6614,
      "step": 8406
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.699370603480193e-06,
      "loss": 1.2493,
      "step": 8407
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.69788967049241e-06,
      "loss": 1.3994,
      "step": 8408
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.696408737504628e-06,
      "loss": 1.4134,
      "step": 8409
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.694927804516847e-06,
      "loss": 1.1604,
      "step": 8410
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.693446871529065e-06,
      "loss": 1.1719,
      "step": 8411
    },
    {
      "epoch": 1.85,
      "learning_rate": 7.691965938541282e-06,
      "loss": 1.1264,
      "step": 8412
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.6904850055535e-06,
      "loss": 1.3425,
      "step": 8413
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.689004072565717e-06,
      "loss": 1.2303,
      "step": 8414
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.687523139577935e-06,
      "loss": 1.3086,
      "step": 8415
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.686042206590152e-06,
      "loss": 1.2866,
      "step": 8416
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.68456127360237e-06,
      "loss": 1.223,
      "step": 8417
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.683080340614589e-06,
      "loss": 1.3358,
      "step": 8418
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.681599407626805e-06,
      "loss": 1.4638,
      "step": 8419
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.680118474639024e-06,
      "loss": 1.6375,
      "step": 8420
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.67863754165124e-06,
      "loss": 1.4036,
      "step": 8421
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.677156608663459e-06,
      "loss": 1.358,
      "step": 8422
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.675675675675676e-06,
      "loss": 1.4815,
      "step": 8423
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.674194742687894e-06,
      "loss": 1.4895,
      "step": 8424
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.672713809700112e-06,
      "loss": 1.3988,
      "step": 8425
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.671232876712329e-06,
      "loss": 1.3174,
      "step": 8426
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.669751943724547e-06,
      "loss": 1.3891,
      "step": 8427
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.668271010736766e-06,
      "loss": 1.2972,
      "step": 8428
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.666790077748982e-06,
      "loss": 1.1857,
      "step": 8429
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.6653091447612e-06,
      "loss": 1.1836,
      "step": 8430
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.663828211773418e-06,
      "loss": 1.1489,
      "step": 8431
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.662347278785636e-06,
      "loss": 1.3065,
      "step": 8432
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.660866345797853e-06,
      "loss": 1.4087,
      "step": 8433
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.659385412810071e-06,
      "loss": 1.2394,
      "step": 8434
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.657904479822288e-06,
      "loss": 1.3953,
      "step": 8435
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.656423546834506e-06,
      "loss": 1.2823,
      "step": 8436
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.654942613846723e-06,
      "loss": 1.2422,
      "step": 8437
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.653461680858941e-06,
      "loss": 1.2893,
      "step": 8438
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.65198074787116e-06,
      "loss": 1.2329,
      "step": 8439
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.650499814883378e-06,
      "loss": 1.1775,
      "step": 8440
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.649018881895595e-06,
      "loss": 1.3013,
      "step": 8441
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.647537948907813e-06,
      "loss": 1.269,
      "step": 8442
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.64605701592003e-06,
      "loss": 1.3205,
      "step": 8443
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.644576082932248e-06,
      "loss": 1.099,
      "step": 8444
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.643095149944467e-06,
      "loss": 1.12,
      "step": 8445
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.641614216956683e-06,
      "loss": 1.4199,
      "step": 8446
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.640133283968902e-06,
      "loss": 1.2833,
      "step": 8447
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.638652350981118e-06,
      "loss": 1.3995,
      "step": 8448
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.637171417993337e-06,
      "loss": 1.2177,
      "step": 8449
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.635690485005555e-06,
      "loss": 1.1594,
      "step": 8450
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.634209552017772e-06,
      "loss": 1.2388,
      "step": 8451
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.63272861902999e-06,
      "loss": 1.376,
      "step": 8452
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.631247686042207e-06,
      "loss": 1.1204,
      "step": 8453
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.629766753054425e-06,
      "loss": 1.0294,
      "step": 8454
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.628285820066642e-06,
      "loss": 1.3689,
      "step": 8455
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.6268048870788605e-06,
      "loss": 1.2493,
      "step": 8456
    },
    {
      "epoch": 1.86,
      "learning_rate": 7.625323954091078e-06,
      "loss": 1.2426,
      "step": 8457
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.6238430211032955e-06,
      "loss": 1.2271,
      "step": 8458
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.622362088115514e-06,
      "loss": 1.3352,
      "step": 8459
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.6208811551277306e-06,
      "loss": 1.3783,
      "step": 8460
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.619400222139949e-06,
      "loss": 1.4249,
      "step": 8461
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.617919289152167e-06,
      "loss": 1.3233,
      "step": 8462
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.616438356164384e-06,
      "loss": 1.1578,
      "step": 8463
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.614957423176602e-06,
      "loss": 1.4349,
      "step": 8464
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.613476490188819e-06,
      "loss": 1.087,
      "step": 8465
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.611995557201037e-06,
      "loss": 1.3888,
      "step": 8466
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.610514624213256e-06,
      "loss": 1.1997,
      "step": 8467
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.609033691225472e-06,
      "loss": 1.4681,
      "step": 8468
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.607552758237691e-06,
      "loss": 1.2033,
      "step": 8469
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.606071825249907e-06,
      "loss": 1.2283,
      "step": 8470
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.604590892262126e-06,
      "loss": 1.2052,
      "step": 8471
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.603109959274343e-06,
      "loss": 1.1562,
      "step": 8472
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.601629026286561e-06,
      "loss": 1.2373,
      "step": 8473
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.600148093298779e-06,
      "loss": 1.2431,
      "step": 8474
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.598667160310997e-06,
      "loss": 1.3164,
      "step": 8475
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.597186227323214e-06,
      "loss": 1.3907,
      "step": 8476
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.595705294335432e-06,
      "loss": 1.3935,
      "step": 8477
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.594224361347649e-06,
      "loss": 1.156,
      "step": 8478
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.592743428359868e-06,
      "loss": 1.2668,
      "step": 8479
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.591262495372085e-06,
      "loss": 1.1646,
      "step": 8480
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.589781562384303e-06,
      "loss": 1.238,
      "step": 8481
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.58830062939652e-06,
      "loss": 1.3749,
      "step": 8482
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.5868196964087385e-06,
      "loss": 1.385,
      "step": 8483
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.585338763420956e-06,
      "loss": 1.2608,
      "step": 8484
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.5838578304331736e-06,
      "loss": 1.3894,
      "step": 8485
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.582376897445392e-06,
      "loss": 1.1523,
      "step": 8486
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.580895964457609e-06,
      "loss": 1.1801,
      "step": 8487
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.579415031469827e-06,
      "loss": 1.2586,
      "step": 8488
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.577934098482044e-06,
      "loss": 1.2723,
      "step": 8489
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.576453165494262e-06,
      "loss": 1.1522,
      "step": 8490
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.57497223250648e-06,
      "loss": 1.167,
      "step": 8491
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.573491299518697e-06,
      "loss": 1.3181,
      "step": 8492
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.572010366530915e-06,
      "loss": 1.3796,
      "step": 8493
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.570529433543132e-06,
      "loss": 1.1462,
      "step": 8494
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.56904850055535e-06,
      "loss": 1.2156,
      "step": 8495
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.567567567567569e-06,
      "loss": 1.0866,
      "step": 8496
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.5660866345797855e-06,
      "loss": 1.3415,
      "step": 8497
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.564605701592004e-06,
      "loss": 1.1574,
      "step": 8498
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.5631247686042205e-06,
      "loss": 1.2375,
      "step": 8499
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.561643835616439e-06,
      "loss": 1.1772,
      "step": 8500
    },
    {
      "epoch": 1.87,
      "eval_loss": NaN,
      "eval_runtime": 4.3757,
      "eval_samples_per_second": 352.857,
      "eval_steps_per_second": 44.107,
      "step": 8500
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.560162902628657e-06,
      "loss": 1.0373,
      "step": 8501
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.558681969640874e-06,
      "loss": 1.2635,
      "step": 8502
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.557201036653092e-06,
      "loss": 1.2185,
      "step": 8503
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.55572010366531e-06,
      "loss": 1.5032,
      "step": 8504
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.554239170677527e-06,
      "loss": 1.486,
      "step": 8505
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.552758237689746e-06,
      "loss": 1.2388,
      "step": 8506
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.551277304701963e-06,
      "loss": 1.2757,
      "step": 8507
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.549796371714181e-06,
      "loss": 1.0261,
      "step": 8508
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.548315438726398e-06,
      "loss": 1.2536,
      "step": 8509
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.546834505738616e-06,
      "loss": 1.3784,
      "step": 8510
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.545353572750833e-06,
      "loss": 1.0587,
      "step": 8511
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.543872639763052e-06,
      "loss": 1.4424,
      "step": 8512
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.542391706775269e-06,
      "loss": 1.3081,
      "step": 8513
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.540910773787487e-06,
      "loss": 1.2326,
      "step": 8514
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.539429840799705e-06,
      "loss": 1.403,
      "step": 8515
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.537948907811922e-06,
      "loss": 1.1541,
      "step": 8516
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.53646797482414e-06,
      "loss": 1.2966,
      "step": 8517
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.534987041836358e-06,
      "loss": 1.3753,
      "step": 8518
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.533506108848575e-06,
      "loss": 1.2904,
      "step": 8519
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.532025175860793e-06,
      "loss": 1.1916,
      "step": 8520
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.53054424287301e-06,
      "loss": 1.3655,
      "step": 8521
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.5290633098852285e-06,
      "loss": 1.2161,
      "step": 8522
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.527582376897447e-06,
      "loss": 1.3006,
      "step": 8523
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.5261014439096635e-06,
      "loss": 1.5388,
      "step": 8524
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.524620510921882e-06,
      "loss": 1.1224,
      "step": 8525
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.5231395779340985e-06,
      "loss": 1.1193,
      "step": 8526
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.521658644946317e-06,
      "loss": 1.0911,
      "step": 8527
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.5201777119585336e-06,
      "loss": 1.3561,
      "step": 8528
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.518696778970752e-06,
      "loss": 1.3043,
      "step": 8529
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.51721584598297e-06,
      "loss": 1.1839,
      "step": 8530
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.515734912995187e-06,
      "loss": 1.2757,
      "step": 8531
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.514253980007405e-06,
      "loss": 1.3064,
      "step": 8532
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.512773047019623e-06,
      "loss": 1.4069,
      "step": 8533
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.51129211403184e-06,
      "loss": 1.4833,
      "step": 8534
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.509811181044059e-06,
      "loss": 1.0588,
      "step": 8535
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.508330248056276e-06,
      "loss": 1.1821,
      "step": 8536
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.506849315068494e-06,
      "loss": 1.2736,
      "step": 8537
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.505368382080711e-06,
      "loss": 1.2683,
      "step": 8538
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.50388744909293e-06,
      "loss": 1.3718,
      "step": 8539
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.502406516105147e-06,
      "loss": 1.2452,
      "step": 8540
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.500925583117365e-06,
      "loss": 1.3647,
      "step": 8541
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.499444650129582e-06,
      "loss": 1.317,
      "step": 8542
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.4979637171418e-06,
      "loss": 1.2321,
      "step": 8543
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.496482784154018e-06,
      "loss": 1.2497,
      "step": 8544
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.495001851166235e-06,
      "loss": 1.3816,
      "step": 8545
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.493520918178453e-06,
      "loss": 1.1746,
      "step": 8546
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.4920399851906715e-06,
      "loss": 1.4256,
      "step": 8547
    },
    {
      "epoch": 1.88,
      "learning_rate": 7.490559052202888e-06,
      "loss": 1.2576,
      "step": 8548
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4890781192151065e-06,
      "loss": 1.3221,
      "step": 8549
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.487597186227323e-06,
      "loss": 1.3722,
      "step": 8550
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4861162532395415e-06,
      "loss": 1.2216,
      "step": 8551
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.48463532025176e-06,
      "loss": 1.1577,
      "step": 8552
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4831543872639766e-06,
      "loss": 1.3992,
      "step": 8553
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.481673454276195e-06,
      "loss": 1.1487,
      "step": 8554
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.480192521288412e-06,
      "loss": 1.2799,
      "step": 8555
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.47871158830063e-06,
      "loss": 1.2872,
      "step": 8556
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.477230655312848e-06,
      "loss": 1.342,
      "step": 8557
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.475749722325065e-06,
      "loss": 1.3494,
      "step": 8558
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.474268789337283e-06,
      "loss": 1.162,
      "step": 8559
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4727878563495e-06,
      "loss": 1.3141,
      "step": 8560
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.471306923361718e-06,
      "loss": 1.2664,
      "step": 8561
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.469825990373937e-06,
      "loss": 1.3733,
      "step": 8562
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4683450573861534e-06,
      "loss": 1.2113,
      "step": 8563
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.466864124398372e-06,
      "loss": 1.1412,
      "step": 8564
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.465383191410589e-06,
      "loss": 1.2136,
      "step": 8565
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.463902258422807e-06,
      "loss": 1.4592,
      "step": 8566
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.462421325435024e-06,
      "loss": 1.2803,
      "step": 8567
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.460940392447243e-06,
      "loss": 0.9662,
      "step": 8568
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.45945945945946e-06,
      "loss": 1.3481,
      "step": 8569
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.457978526471678e-06,
      "loss": 1.2492,
      "step": 8570
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.456497593483896e-06,
      "loss": 1.1108,
      "step": 8571
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.455016660496113e-06,
      "loss": 1.1482,
      "step": 8572
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.453535727508331e-06,
      "loss": 1.457,
      "step": 8573
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.452054794520549e-06,
      "loss": 1.218,
      "step": 8574
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.450573861532766e-06,
      "loss": 1.3431,
      "step": 8575
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4490929285449845e-06,
      "loss": 1.0254,
      "step": 8576
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.447611995557201e-06,
      "loss": 1.2801,
      "step": 8577
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4461310625694196e-06,
      "loss": 1.2991,
      "step": 8578
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.444650129581638e-06,
      "loss": 1.3817,
      "step": 8579
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.443169196593855e-06,
      "loss": 1.1515,
      "step": 8580
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.441688263606073e-06,
      "loss": 1.4634,
      "step": 8581
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.44020733061829e-06,
      "loss": 1.3479,
      "step": 8582
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.438726397630508e-06,
      "loss": 1.1175,
      "step": 8583
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.437245464642725e-06,
      "loss": 1.4472,
      "step": 8584
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.435764531654943e-06,
      "loss": 1.2259,
      "step": 8585
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.434283598667161e-06,
      "loss": 1.103,
      "step": 8586
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.432802665679378e-06,
      "loss": 1.5935,
      "step": 8587
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4313217326915964e-06,
      "loss": 1.3323,
      "step": 8588
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.429840799703814e-06,
      "loss": 1.2674,
      "step": 8589
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4283598667160315e-06,
      "loss": 1.1537,
      "step": 8590
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.42687893372825e-06,
      "loss": 1.3315,
      "step": 8591
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.4253980007404665e-06,
      "loss": 1.3519,
      "step": 8592
    },
    {
      "epoch": 1.89,
      "learning_rate": 7.423917067752685e-06,
      "loss": 1.3784,
      "step": 8593
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.422436134764902e-06,
      "loss": 1.3134,
      "step": 8594
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.42095520177712e-06,
      "loss": 1.2983,
      "step": 8595
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.419474268789338e-06,
      "loss": 1.1947,
      "step": 8596
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.417993335801556e-06,
      "loss": 1.2505,
      "step": 8597
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.416512402813773e-06,
      "loss": 1.2346,
      "step": 8598
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.415031469825991e-06,
      "loss": 1.426,
      "step": 8599
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.413550536838209e-06,
      "loss": 1.3863,
      "step": 8600
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.412069603850426e-06,
      "loss": 1.1922,
      "step": 8601
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.410588670862644e-06,
      "loss": 1.0743,
      "step": 8602
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.4091077378748626e-06,
      "loss": 1.3362,
      "step": 8603
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.407626804887079e-06,
      "loss": 1.3085,
      "step": 8604
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.406145871899298e-06,
      "loss": 1.3801,
      "step": 8605
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.404664938911514e-06,
      "loss": 1.3361,
      "step": 8606
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.403184005923733e-06,
      "loss": 1.0333,
      "step": 8607
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.401703072935951e-06,
      "loss": 1.2891,
      "step": 8608
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.400222139948168e-06,
      "loss": 1.5272,
      "step": 8609
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.398741206960386e-06,
      "loss": 1.2433,
      "step": 8610
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.397260273972603e-06,
      "loss": 1.4001,
      "step": 8611
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.395779340984821e-06,
      "loss": 1.2075,
      "step": 8612
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.3942984079970394e-06,
      "loss": 1.2231,
      "step": 8613
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.392817475009256e-06,
      "loss": 1.3756,
      "step": 8614
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.3913365420214745e-06,
      "loss": 1.3313,
      "step": 8615
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.389855609033691e-06,
      "loss": 1.3677,
      "step": 8616
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.3883746760459095e-06,
      "loss": 1.3439,
      "step": 8617
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.386893743058127e-06,
      "loss": 1.2283,
      "step": 8618
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.3854128100703445e-06,
      "loss": 1.2031,
      "step": 8619
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.383931877082563e-06,
      "loss": 1.5561,
      "step": 8620
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.38245094409478e-06,
      "loss": 1.2706,
      "step": 8621
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.380970011106998e-06,
      "loss": 1.3957,
      "step": 8622
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.3794890781192155e-06,
      "loss": 1.4583,
      "step": 8623
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.378008145131433e-06,
      "loss": 1.182,
      "step": 8624
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.376527212143651e-06,
      "loss": 1.3211,
      "step": 8625
    },
    {
      "epoch": 1.9,
      "eval_loss": NaN,
      "eval_runtime": 4.3732,
      "eval_samples_per_second": 353.06,
      "eval_steps_per_second": 44.132,
      "step": 8625
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.375046279155869e-06,
      "loss": 1.3246,
      "step": 8626
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.373565346168086e-06,
      "loss": 1.4127,
      "step": 8627
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.372084413180304e-06,
      "loss": 1.0697,
      "step": 8628
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.370603480192522e-06,
      "loss": 1.3814,
      "step": 8629
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.36912254720474e-06,
      "loss": 1.399,
      "step": 8630
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.367641614216957e-06,
      "loss": 1.3365,
      "step": 8631
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.366160681229176e-06,
      "loss": 1.2598,
      "step": 8632
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.364679748241392e-06,
      "loss": 1.2574,
      "step": 8633
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.363198815253611e-06,
      "loss": 1.5819,
      "step": 8634
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.361717882265829e-06,
      "loss": 1.3461,
      "step": 8635
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.360236949278046e-06,
      "loss": 1.3451,
      "step": 8636
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.358756016290264e-06,
      "loss": 1.0803,
      "step": 8637
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.357275083302481e-06,
      "loss": 1.2627,
      "step": 8638
    },
    {
      "epoch": 1.9,
      "learning_rate": 7.355794150314699e-06,
      "loss": 1.4788,
      "step": 8639
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.354313217326916e-06,
      "loss": 1.1577,
      "step": 8640
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.352832284339134e-06,
      "loss": 1.3092,
      "step": 8641
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.3513513513513525e-06,
      "loss": 1.2415,
      "step": 8642
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.349870418363569e-06,
      "loss": 1.0864,
      "step": 8643
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.3483894853757875e-06,
      "loss": 1.4965,
      "step": 8644
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.346908552388004e-06,
      "loss": 1.1423,
      "step": 8645
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.3454276194002226e-06,
      "loss": 1.4906,
      "step": 8646
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.343946686412441e-06,
      "loss": 1.032,
      "step": 8647
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.342465753424658e-06,
      "loss": 1.2556,
      "step": 8648
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.340984820436876e-06,
      "loss": 1.0862,
      "step": 8649
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.3395038874490935e-06,
      "loss": 1.2126,
      "step": 8650
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.338022954461311e-06,
      "loss": 1.3654,
      "step": 8651
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.336542021473529e-06,
      "loss": 1.3309,
      "step": 8652
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.335061088485747e-06,
      "loss": 1.2532,
      "step": 8653
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.333580155497964e-06,
      "loss": 1.6514,
      "step": 8654
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.332099222510182e-06,
      "loss": 1.3718,
      "step": 8655
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.3306182895223994e-06,
      "loss": 1.3402,
      "step": 8656
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.329137356534617e-06,
      "loss": 1.2968,
      "step": 8657
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.327656423546835e-06,
      "loss": 1.3999,
      "step": 8658
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.326175490559053e-06,
      "loss": 1.2541,
      "step": 8659
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.32469455757127e-06,
      "loss": 1.3252,
      "step": 8660
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.323213624583489e-06,
      "loss": 1.3094,
      "step": 8661
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.321732691595705e-06,
      "loss": 1.332,
      "step": 8662
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.320251758607924e-06,
      "loss": 1.3304,
      "step": 8663
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.318770825620142e-06,
      "loss": 1.323,
      "step": 8664
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.317289892632359e-06,
      "loss": 1.2144,
      "step": 8665
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.315808959644577e-06,
      "loss": 1.2742,
      "step": 8666
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.314328026656794e-06,
      "loss": 1.6256,
      "step": 8667
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.312847093669012e-06,
      "loss": 1.2197,
      "step": 8668
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.3113661606812305e-06,
      "loss": 1.3298,
      "step": 8669
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.309885227693447e-06,
      "loss": 1.3244,
      "step": 8670
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.308404294705666e-06,
      "loss": 1.2777,
      "step": 8671
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.306923361717882e-06,
      "loss": 1.1622,
      "step": 8672
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.305442428730101e-06,
      "loss": 1.1669,
      "step": 8673
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.303961495742318e-06,
      "loss": 1.2772,
      "step": 8674
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.302480562754536e-06,
      "loss": 1.3702,
      "step": 8675
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.300999629766754e-06,
      "loss": 1.1701,
      "step": 8676
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.299518696778971e-06,
      "loss": 1.263,
      "step": 8677
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.298037763791189e-06,
      "loss": 1.252,
      "step": 8678
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.2965568308034066e-06,
      "loss": 1.1926,
      "step": 8679
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.295075897815624e-06,
      "loss": 1.3006,
      "step": 8680
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.2935949648278424e-06,
      "loss": 1.3381,
      "step": 8681
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.29211403184006e-06,
      "loss": 1.4221,
      "step": 8682
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.2906330988522775e-06,
      "loss": 1.2371,
      "step": 8683
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.289152165864495e-06,
      "loss": 1.1225,
      "step": 8684
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.287671232876713e-06,
      "loss": 1.3578,
      "step": 8685
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.286190299888931e-06,
      "loss": 1.1687,
      "step": 8686
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.284709366901148e-06,
      "loss": 1.3515,
      "step": 8687
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.283228433913366e-06,
      "loss": 1.4646,
      "step": 8688
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.2817475009255834e-06,
      "loss": 1.3138,
      "step": 8689
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.280266567937802e-06,
      "loss": 1.2718,
      "step": 8690
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.2787856349500185e-06,
      "loss": 1.3681,
      "step": 8691
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.277304701962237e-06,
      "loss": 1.2841,
      "step": 8692
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.275823768974455e-06,
      "loss": 1.1963,
      "step": 8693
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.274342835986672e-06,
      "loss": 1.5645,
      "step": 8694
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.27286190299889e-06,
      "loss": 1.4167,
      "step": 8695
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.271380970011107e-06,
      "loss": 1.2594,
      "step": 8696
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.269900037023325e-06,
      "loss": 1.4157,
      "step": 8697
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.268419104035544e-06,
      "loss": 1.2432,
      "step": 8698
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.26693817104776e-06,
      "loss": 1.1114,
      "step": 8699
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.265457238059979e-06,
      "loss": 1.4352,
      "step": 8700
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.263976305072195e-06,
      "loss": 1.5092,
      "step": 8701
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.262495372084414e-06,
      "loss": 1.4214,
      "step": 8702
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.261014439096632e-06,
      "loss": 1.2371,
      "step": 8703
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.259533506108849e-06,
      "loss": 1.2774,
      "step": 8704
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.258052573121067e-06,
      "loss": 1.334,
      "step": 8705
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.256571640133285e-06,
      "loss": 1.3602,
      "step": 8706
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.255090707145502e-06,
      "loss": 1.7885,
      "step": 8707
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.2536097741577205e-06,
      "loss": 1.2448,
      "step": 8708
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.252128841169937e-06,
      "loss": 1.1183,
      "step": 8709
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.2506479081821555e-06,
      "loss": 1.2762,
      "step": 8710
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.249166975194373e-06,
      "loss": 1.3702,
      "step": 8711
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.2476860422065905e-06,
      "loss": 1.2501,
      "step": 8712
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.246205109218808e-06,
      "loss": 1.4818,
      "step": 8713
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.2447241762310264e-06,
      "loss": 1.1649,
      "step": 8714
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.243243243243244e-06,
      "loss": 1.3747,
      "step": 8715
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.2417623102554615e-06,
      "loss": 1.4611,
      "step": 8716
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.24028137726768e-06,
      "loss": 1.1952,
      "step": 8717
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.2388004442798965e-06,
      "loss": 1.1472,
      "step": 8718
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.237319511292115e-06,
      "loss": 1.1979,
      "step": 8719
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.235838578304332e-06,
      "loss": 1.3646,
      "step": 8720
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.23435764531655e-06,
      "loss": 1.2218,
      "step": 8721
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.232876712328768e-06,
      "loss": 1.0877,
      "step": 8722
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.231395779340985e-06,
      "loss": 1.3862,
      "step": 8723
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.229914846353203e-06,
      "loss": 1.1664,
      "step": 8724
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.228433913365422e-06,
      "loss": 1.2946,
      "step": 8725
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.226952980377638e-06,
      "loss": 1.3828,
      "step": 8726
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.225472047389857e-06,
      "loss": 1.2584,
      "step": 8727
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.223991114402073e-06,
      "loss": 1.2751,
      "step": 8728
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.222510181414292e-06,
      "loss": 1.4523,
      "step": 8729
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.221029248426508e-06,
      "loss": 1.2169,
      "step": 8730
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.219548315438727e-06,
      "loss": 1.3552,
      "step": 8731
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.218067382450945e-06,
      "loss": 1.2137,
      "step": 8732
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.216586449463162e-06,
      "loss": 1.4738,
      "step": 8733
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.21510551647538e-06,
      "loss": 1.3591,
      "step": 8734
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.213624583487598e-06,
      "loss": 1.4006,
      "step": 8735
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.212143650499815e-06,
      "loss": 1.0931,
      "step": 8736
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.2106627175120336e-06,
      "loss": 1.3218,
      "step": 8737
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.209181784524251e-06,
      "loss": 1.0086,
      "step": 8738
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.207700851536469e-06,
      "loss": 1.2715,
      "step": 8739
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.206219918548686e-06,
      "loss": 1.2308,
      "step": 8740
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.204738985560904e-06,
      "loss": 1.4353,
      "step": 8741
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.203258052573122e-06,
      "loss": 1.4132,
      "step": 8742
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.2017771195853395e-06,
      "loss": 1.2525,
      "step": 8743
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.200296186597557e-06,
      "loss": 1.2143,
      "step": 8744
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.1988152536097745e-06,
      "loss": 1.2523,
      "step": 8745
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.197334320621993e-06,
      "loss": 1.1531,
      "step": 8746
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.1958533876342096e-06,
      "loss": 1.1997,
      "step": 8747
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.194372454646428e-06,
      "loss": 1.309,
      "step": 8748
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.192891521658646e-06,
      "loss": 1.277,
      "step": 8749
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.191410588670863e-06,
      "loss": 1.4378,
      "step": 8750
    },
    {
      "epoch": 1.93,
      "eval_loss": NaN,
      "eval_runtime": 4.3761,
      "eval_samples_per_second": 352.823,
      "eval_steps_per_second": 44.103,
      "step": 8750
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.189929655683081e-06,
      "loss": 1.4183,
      "step": 8751
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.188448722695298e-06,
      "loss": 1.1726,
      "step": 8752
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.186967789707516e-06,
      "loss": 1.2779,
      "step": 8753
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.185486856719735e-06,
      "loss": 1.2722,
      "step": 8754
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.184005923731951e-06,
      "loss": 1.2361,
      "step": 8755
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.18252499074417e-06,
      "loss": 1.4069,
      "step": 8756
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.1810440577563864e-06,
      "loss": 1.2064,
      "step": 8757
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.179563124768605e-06,
      "loss": 1.3002,
      "step": 8758
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.178082191780823e-06,
      "loss": 1.2971,
      "step": 8759
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.17660125879304e-06,
      "loss": 1.2552,
      "step": 8760
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.175120325805258e-06,
      "loss": 1.3103,
      "step": 8761
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.173639392817475e-06,
      "loss": 1.2758,
      "step": 8762
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.172158459829693e-06,
      "loss": 1.3401,
      "step": 8763
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.170677526841912e-06,
      "loss": 1.2803,
      "step": 8764
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.169196593854128e-06,
      "loss": 1.275,
      "step": 8765
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.167715660866347e-06,
      "loss": 1.2969,
      "step": 8766
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.166234727878564e-06,
      "loss": 1.3515,
      "step": 8767
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.164753794890782e-06,
      "loss": 1.2047,
      "step": 8768
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.163272861902999e-06,
      "loss": 1.3146,
      "step": 8769
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.1617919289152175e-06,
      "loss": 1.343,
      "step": 8770
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.160310995927435e-06,
      "loss": 1.4641,
      "step": 8771
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.1588300629396526e-06,
      "loss": 1.1247,
      "step": 8772
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.15734912995187e-06,
      "loss": 1.3483,
      "step": 8773
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.155868196964088e-06,
      "loss": 1.4992,
      "step": 8774
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.154387263976306e-06,
      "loss": 1.4686,
      "step": 8775
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.1529063309885235e-06,
      "loss": 1.4287,
      "step": 8776
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.151425398000741e-06,
      "loss": 1.4815,
      "step": 8777
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.149944465012959e-06,
      "loss": 1.3054,
      "step": 8778
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.148463532025176e-06,
      "loss": 1.3765,
      "step": 8779
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.146982599037394e-06,
      "loss": 1.1497,
      "step": 8780
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.145501666049613e-06,
      "loss": 1.3558,
      "step": 8781
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.1440207330618294e-06,
      "loss": 1.0279,
      "step": 8782
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.142539800074048e-06,
      "loss": 1.4856,
      "step": 8783
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.1410588670862645e-06,
      "loss": 1.2191,
      "step": 8784
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.139577934098483e-06,
      "loss": 1.1261,
      "step": 8785
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.1380970011106995e-06,
      "loss": 1.1758,
      "step": 8786
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.136616068122918e-06,
      "loss": 1.271,
      "step": 8787
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.135135135135136e-06,
      "loss": 1.2359,
      "step": 8788
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.133654202147353e-06,
      "loss": 1.2296,
      "step": 8789
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.132173269159571e-06,
      "loss": 1.4129,
      "step": 8790
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.130692336171788e-06,
      "loss": 1.2808,
      "step": 8791
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.129211403184006e-06,
      "loss": 1.1838,
      "step": 8792
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.127730470196225e-06,
      "loss": 1.2489,
      "step": 8793
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.126249537208441e-06,
      "loss": 1.1599,
      "step": 8794
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.12476860422066e-06,
      "loss": 1.3297,
      "step": 8795
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.123287671232877e-06,
      "loss": 1.3104,
      "step": 8796
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.121806738245095e-06,
      "loss": 1.1161,
      "step": 8797
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.120325805257313e-06,
      "loss": 1.1781,
      "step": 8798
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.118844872269531e-06,
      "loss": 1.2804,
      "step": 8799
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.117363939281748e-06,
      "loss": 1.2158,
      "step": 8800
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.115883006293966e-06,
      "loss": 1.1191,
      "step": 8801
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.114402073306184e-06,
      "loss": 1.2226,
      "step": 8802
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.112921140318401e-06,
      "loss": 1.3424,
      "step": 8803
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.111440207330619e-06,
      "loss": 1.1798,
      "step": 8804
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.1099592743428366e-06,
      "loss": 1.2241,
      "step": 8805
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.108478341355054e-06,
      "loss": 1.2833,
      "step": 8806
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.1069974083672724e-06,
      "loss": 1.2635,
      "step": 8807
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.105516475379489e-06,
      "loss": 1.359,
      "step": 8808
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.1040355423917075e-06,
      "loss": 1.0405,
      "step": 8809
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.102554609403926e-06,
      "loss": 1.2958,
      "step": 8810
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.1010736764161425e-06,
      "loss": 1.204,
      "step": 8811
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.099592743428361e-06,
      "loss": 1.2204,
      "step": 8812
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.0981118104405775e-06,
      "loss": 1.4062,
      "step": 8813
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.096630877452796e-06,
      "loss": 1.107,
      "step": 8814
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.095149944465014e-06,
      "loss": 1.3116,
      "step": 8815
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.093669011477231e-06,
      "loss": 1.3983,
      "step": 8816
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.092188078489449e-06,
      "loss": 1.4909,
      "step": 8817
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.090707145501666e-06,
      "loss": 1.2028,
      "step": 8818
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.089226212513884e-06,
      "loss": 1.2829,
      "step": 8819
    },
    {
      "epoch": 1.94,
      "learning_rate": 7.087745279526102e-06,
      "loss": 1.2959,
      "step": 8820
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.086264346538319e-06,
      "loss": 1.2753,
      "step": 8821
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.084783413550538e-06,
      "loss": 1.1704,
      "step": 8822
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.083302480562754e-06,
      "loss": 1.2341,
      "step": 8823
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.081821547574973e-06,
      "loss": 1.2967,
      "step": 8824
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.08034061458719e-06,
      "loss": 1.2685,
      "step": 8825
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.078859681599408e-06,
      "loss": 1.287,
      "step": 8826
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.077378748611626e-06,
      "loss": 1.4151,
      "step": 8827
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.075897815623844e-06,
      "loss": 1.3978,
      "step": 8828
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.074416882636061e-06,
      "loss": 1.3293,
      "step": 8829
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.072935949648279e-06,
      "loss": 1.3094,
      "step": 8830
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.071455016660497e-06,
      "loss": 1.2868,
      "step": 8831
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.069974083672715e-06,
      "loss": 1.2039,
      "step": 8832
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.068493150684932e-06,
      "loss": 1.3565,
      "step": 8833
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.0670122176971505e-06,
      "loss": 1.2541,
      "step": 8834
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.065531284709367e-06,
      "loss": 1.2211,
      "step": 8835
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.0640503517215855e-06,
      "loss": 1.3614,
      "step": 8836
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.062569418733803e-06,
      "loss": 1.1169,
      "step": 8837
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.0610884857460205e-06,
      "loss": 1.2109,
      "step": 8838
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.059607552758239e-06,
      "loss": 1.1691,
      "step": 8839
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.058126619770456e-06,
      "loss": 1.2196,
      "step": 8840
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.056645686782674e-06,
      "loss": 1.5027,
      "step": 8841
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.055164753794891e-06,
      "loss": 1.2095,
      "step": 8842
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.053683820807109e-06,
      "loss": 1.2478,
      "step": 8843
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.052202887819327e-06,
      "loss": 1.711,
      "step": 8844
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.050721954831544e-06,
      "loss": 1.4025,
      "step": 8845
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.049241021843762e-06,
      "loss": 1.3191,
      "step": 8846
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.047760088855979e-06,
      "loss": 1.4757,
      "step": 8847
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.046279155868197e-06,
      "loss": 1.3282,
      "step": 8848
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.044798222880416e-06,
      "loss": 1.5031,
      "step": 8849
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.0433172898926324e-06,
      "loss": 1.2967,
      "step": 8850
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.041836356904851e-06,
      "loss": 1.5604,
      "step": 8851
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.040355423917068e-06,
      "loss": 1.3472,
      "step": 8852
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.038874490929286e-06,
      "loss": 1.1583,
      "step": 8853
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.037393557941504e-06,
      "loss": 1.1688,
      "step": 8854
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.035912624953721e-06,
      "loss": 1.4258,
      "step": 8855
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.034431691965939e-06,
      "loss": 1.5051,
      "step": 8856
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.032950758978157e-06,
      "loss": 1.2006,
      "step": 8857
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.031469825990374e-06,
      "loss": 1.3569,
      "step": 8858
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.029988893002592e-06,
      "loss": 1.331,
      "step": 8859
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.02850796001481e-06,
      "loss": 1.2928,
      "step": 8860
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.027027027027028e-06,
      "loss": 1.0012,
      "step": 8861
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.025546094039245e-06,
      "loss": 1.4327,
      "step": 8862
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.0240651610514636e-06,
      "loss": 1.3073,
      "step": 8863
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.02258422806368e-06,
      "loss": 1.1649,
      "step": 8864
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.021103295075899e-06,
      "loss": 1.198,
      "step": 8865
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.019622362088117e-06,
      "loss": 1.4142,
      "step": 8866
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.018141429100334e-06,
      "loss": 1.4278,
      "step": 8867
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.016660496112552e-06,
      "loss": 1.2356,
      "step": 8868
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.015179563124769e-06,
      "loss": 1.136,
      "step": 8869
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.013698630136987e-06,
      "loss": 1.394,
      "step": 8870
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.012217697149205e-06,
      "loss": 1.3911,
      "step": 8871
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.010736764161422e-06,
      "loss": 1.2494,
      "step": 8872
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.00925583117364e-06,
      "loss": 1.5333,
      "step": 8873
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.007774898185857e-06,
      "loss": 1.3026,
      "step": 8874
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.0062939651980754e-06,
      "loss": 1.2672,
      "step": 8875
    },
    {
      "epoch": 1.96,
      "eval_loss": NaN,
      "eval_runtime": 4.3778,
      "eval_samples_per_second": 352.686,
      "eval_steps_per_second": 44.086,
      "step": 8875
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.004813032210292e-06,
      "loss": 1.3656,
      "step": 8876
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.0033320992225105e-06,
      "loss": 1.2797,
      "step": 8877
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.001851166234729e-06,
      "loss": 1.5677,
      "step": 8878
    },
    {
      "epoch": 1.96,
      "learning_rate": 7.0003702332469455e-06,
      "loss": 1.3219,
      "step": 8879
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.998889300259164e-06,
      "loss": 1.1059,
      "step": 8880
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.997408367271381e-06,
      "loss": 1.0976,
      "step": 8881
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.995927434283599e-06,
      "loss": 1.2614,
      "step": 8882
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.994446501295817e-06,
      "loss": 1.3385,
      "step": 8883
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.992965568308035e-06,
      "loss": 1.4185,
      "step": 8884
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.991484635320252e-06,
      "loss": 1.2883,
      "step": 8885
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.99000370233247e-06,
      "loss": 1.0584,
      "step": 8886
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.988522769344687e-06,
      "loss": 1.3013,
      "step": 8887
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.987041836356906e-06,
      "loss": 1.2521,
      "step": 8888
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.985560903369123e-06,
      "loss": 0.9295,
      "step": 8889
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.984079970381341e-06,
      "loss": 1.297,
      "step": 8890
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.982599037393558e-06,
      "loss": 1.2739,
      "step": 8891
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.981118104405777e-06,
      "loss": 1.2529,
      "step": 8892
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.979637171417994e-06,
      "loss": 1.4911,
      "step": 8893
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.978156238430212e-06,
      "loss": 1.314,
      "step": 8894
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.97667530544243e-06,
      "loss": 1.2664,
      "step": 8895
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.975194372454647e-06,
      "loss": 1.1449,
      "step": 8896
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.973713439466865e-06,
      "loss": 1.217,
      "step": 8897
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.972232506479082e-06,
      "loss": 1.4903,
      "step": 8898
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.9707515734913e-06,
      "loss": 1.1471,
      "step": 8899
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.9692706405035185e-06,
      "loss": 1.1143,
      "step": 8900
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.967789707515735e-06,
      "loss": 1.408,
      "step": 8901
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.9663087745279535e-06,
      "loss": 1.2281,
      "step": 8902
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.96482784154017e-06,
      "loss": 1.1207,
      "step": 8903
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.9633469085523885e-06,
      "loss": 1.3444,
      "step": 8904
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.961865975564607e-06,
      "loss": 1.2253,
      "step": 8905
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.9603850425768236e-06,
      "loss": 1.2744,
      "step": 8906
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.958904109589042e-06,
      "loss": 1.3119,
      "step": 8907
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.957423176601259e-06,
      "loss": 1.3903,
      "step": 8908
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.955942243613477e-06,
      "loss": 1.0565,
      "step": 8909
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.954461310625695e-06,
      "loss": 1.3815,
      "step": 8910
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.952980377637912e-06,
      "loss": 1.1552,
      "step": 8911
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.95149944465013e-06,
      "loss": 1.1597,
      "step": 8912
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.950018511662348e-06,
      "loss": 1.4111,
      "step": 8913
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.948537578674565e-06,
      "loss": 1.3917,
      "step": 8914
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.947056645686783e-06,
      "loss": 1.2583,
      "step": 8915
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.945575712699001e-06,
      "loss": 1.4176,
      "step": 8916
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.944094779711219e-06,
      "loss": 1.3119,
      "step": 8917
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.942613846723436e-06,
      "loss": 1.3143,
      "step": 8918
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.941132913735654e-06,
      "loss": 1.0631,
      "step": 8919
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.939651980747871e-06,
      "loss": 1.1868,
      "step": 8920
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.93817104776009e-06,
      "loss": 1.403,
      "step": 8921
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.936690114772307e-06,
      "loss": 1.524,
      "step": 8922
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.935209181784525e-06,
      "loss": 1.251,
      "step": 8923
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.933728248796743e-06,
      "loss": 1.4279,
      "step": 8924
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.93224731580896e-06,
      "loss": 1.3293,
      "step": 8925
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.930766382821178e-06,
      "loss": 1.0747,
      "step": 8926
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.9292854498333965e-06,
      "loss": 1.0508,
      "step": 8927
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.927804516845613e-06,
      "loss": 1.1906,
      "step": 8928
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.9263235838578315e-06,
      "loss": 1.535,
      "step": 8929
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.924842650870048e-06,
      "loss": 1.2493,
      "step": 8930
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.9233617178822666e-06,
      "loss": 1.1219,
      "step": 8931
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.921880784894483e-06,
      "loss": 1.7977,
      "step": 8932
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.920399851906702e-06,
      "loss": 1.1392,
      "step": 8933
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.91891891891892e-06,
      "loss": 1.2946,
      "step": 8934
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.917437985931137e-06,
      "loss": 1.2772,
      "step": 8935
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.915957052943355e-06,
      "loss": 1.2891,
      "step": 8936
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.9144761199555725e-06,
      "loss": 1.3789,
      "step": 8937
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.91299518696779e-06,
      "loss": 1.2602,
      "step": 8938
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.911514253980008e-06,
      "loss": 1.2831,
      "step": 8939
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.910033320992225e-06,
      "loss": 1.6181,
      "step": 8940
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.908552388004443e-06,
      "loss": 1.2084,
      "step": 8941
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.907071455016661e-06,
      "loss": 1.3405,
      "step": 8942
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.9055905220288785e-06,
      "loss": 1.1996,
      "step": 8943
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.904109589041097e-06,
      "loss": 1.2016,
      "step": 8944
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.902628656053314e-06,
      "loss": 1.3125,
      "step": 8945
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.901147723065532e-06,
      "loss": 1.6508,
      "step": 8946
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.899666790077749e-06,
      "loss": 1.3469,
      "step": 8947
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.898185857089968e-06,
      "loss": 1.2723,
      "step": 8948
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.896704924102184e-06,
      "loss": 1.3018,
      "step": 8949
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.895223991114403e-06,
      "loss": 1.0344,
      "step": 8950
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.89374305812662e-06,
      "loss": 1.214,
      "step": 8951
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.892262125138838e-06,
      "loss": 1.2777,
      "step": 8952
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.890781192151056e-06,
      "loss": 1.2916,
      "step": 8953
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.889300259163273e-06,
      "loss": 1.3681,
      "step": 8954
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.887819326175491e-06,
      "loss": 1.263,
      "step": 8955
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.8863383931877096e-06,
      "loss": 1.3365,
      "step": 8956
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.884857460199926e-06,
      "loss": 1.2114,
      "step": 8957
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.883376527212145e-06,
      "loss": 0.9967,
      "step": 8958
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.881895594224361e-06,
      "loss": 1.5164,
      "step": 8959
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.88041466123658e-06,
      "loss": 1.3558,
      "step": 8960
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.878933728248798e-06,
      "loss": 1.3698,
      "step": 8961
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.877452795261015e-06,
      "loss": 1.3864,
      "step": 8962
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.875971862273233e-06,
      "loss": 1.3107,
      "step": 8963
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.87449092928545e-06,
      "loss": 1.3436,
      "step": 8964
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.873009996297668e-06,
      "loss": 1.223,
      "step": 8965
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.8715290633098864e-06,
      "loss": 1.1801,
      "step": 8966
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.870048130322103e-06,
      "loss": 1.1509,
      "step": 8967
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.8685671973343215e-06,
      "loss": 1.2904,
      "step": 8968
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.867086264346539e-06,
      "loss": 1.2605,
      "step": 8969
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.8656053313587565e-06,
      "loss": 1.2727,
      "step": 8970
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.864124398370974e-06,
      "loss": 1.2959,
      "step": 8971
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.8626434653831915e-06,
      "loss": 1.3935,
      "step": 8972
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.86116253239541e-06,
      "loss": 1.3264,
      "step": 8973
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.859681599407627e-06,
      "loss": 1.2195,
      "step": 8974
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.858200666419845e-06,
      "loss": 1.0822,
      "step": 8975
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.8567197334320624e-06,
      "loss": 1.0788,
      "step": 8976
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.855238800444281e-06,
      "loss": 1.3065,
      "step": 8977
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.853757867456498e-06,
      "loss": 1.4886,
      "step": 8978
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.852276934468716e-06,
      "loss": 1.2351,
      "step": 8979
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.850796001480934e-06,
      "loss": 1.0922,
      "step": 8980
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.849315068493151e-06,
      "loss": 1.2534,
      "step": 8981
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.847834135505369e-06,
      "loss": 1.1849,
      "step": 8982
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.846353202517587e-06,
      "loss": 1.264,
      "step": 8983
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.844872269529804e-06,
      "loss": 1.7066,
      "step": 8984
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.843391336542023e-06,
      "loss": 1.1619,
      "step": 8985
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.841910403554239e-06,
      "loss": 1.3063,
      "step": 8986
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.840429470566458e-06,
      "loss": 1.0557,
      "step": 8987
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.838948537578674e-06,
      "loss": 1.4285,
      "step": 8988
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.837467604590893e-06,
      "loss": 1.3212,
      "step": 8989
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.835986671603111e-06,
      "loss": 1.1917,
      "step": 8990
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.834505738615328e-06,
      "loss": 1.5168,
      "step": 8991
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.833024805627546e-06,
      "loss": 1.1996,
      "step": 8992
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.831543872639763e-06,
      "loss": 1.6254,
      "step": 8993
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.830062939651981e-06,
      "loss": 1.3259,
      "step": 8994
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.8285820066641995e-06,
      "loss": 1.3229,
      "step": 8995
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.827101073676416e-06,
      "loss": 1.5362,
      "step": 8996
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.8256201406886345e-06,
      "loss": 1.2997,
      "step": 8997
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.824139207700852e-06,
      "loss": 1.2421,
      "step": 8998
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.8226582747130696e-06,
      "loss": 1.224,
      "step": 8999
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.821177341725288e-06,
      "loss": 1.3,
      "step": 9000
    },
    {
      "epoch": 1.98,
      "eval_loss": NaN,
      "eval_runtime": 4.3818,
      "eval_samples_per_second": 352.363,
      "eval_steps_per_second": 44.045,
      "step": 9000
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.8196964087375054e-06,
      "loss": 1.2981,
      "step": 9001
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.818215475749723e-06,
      "loss": 1.2704,
      "step": 9002
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.8167345427619405e-06,
      "loss": 1.5162,
      "step": 9003
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.815253609774158e-06,
      "loss": 1.2621,
      "step": 9004
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.8137726767863755e-06,
      "loss": 1.1542,
      "step": 9005
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.812291743798594e-06,
      "loss": 1.2472,
      "step": 9006
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.810810810810811e-06,
      "loss": 1.5385,
      "step": 9007
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.809329877823029e-06,
      "loss": 1.2105,
      "step": 9008
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.807848944835247e-06,
      "loss": 1.4891,
      "step": 9009
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.806368011847464e-06,
      "loss": 1.357,
      "step": 9010
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.804887078859682e-06,
      "loss": 1.3552,
      "step": 9011
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.803406145871901e-06,
      "loss": 1.1889,
      "step": 9012
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.801925212884117e-06,
      "loss": 1.1462,
      "step": 9013
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.800444279896336e-06,
      "loss": 1.4761,
      "step": 9014
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.798963346908552e-06,
      "loss": 1.5334,
      "step": 9015
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.797482413920771e-06,
      "loss": 1.2398,
      "step": 9016
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.796001480932989e-06,
      "loss": 1.541,
      "step": 9017
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.794520547945206e-06,
      "loss": 1.1852,
      "step": 9018
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.793039614957424e-06,
      "loss": 1.3036,
      "step": 9019
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.791558681969641e-06,
      "loss": 1.3719,
      "step": 9020
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.790077748981859e-06,
      "loss": 1.1658,
      "step": 9021
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.788596815994077e-06,
      "loss": 1.0815,
      "step": 9022
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.787115883006294e-06,
      "loss": 1.2826,
      "step": 9023
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.7856349500185126e-06,
      "loss": 1.3963,
      "step": 9024
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.784154017030729e-06,
      "loss": 1.4338,
      "step": 9025
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.782673084042948e-06,
      "loss": 1.4274,
      "step": 9026
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.781192151055165e-06,
      "loss": 1.3255,
      "step": 9027
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.779711218067383e-06,
      "loss": 1.126,
      "step": 9028
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.778230285079601e-06,
      "loss": 1.4586,
      "step": 9029
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.7767493520918185e-06,
      "loss": 1.2634,
      "step": 9030
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.775268419104036e-06,
      "loss": 1.3174,
      "step": 9031
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.7737874861162536e-06,
      "loss": 1.2948,
      "step": 9032
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.772306553128472e-06,
      "loss": 1.4647,
      "step": 9033
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.7708256201406894e-06,
      "loss": 1.3737,
      "step": 9034
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.769344687152907e-06,
      "loss": 1.2833,
      "step": 9035
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.7678637541651245e-06,
      "loss": 1.2407,
      "step": 9036
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.766382821177342e-06,
      "loss": 1.2541,
      "step": 9037
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.76490188818956e-06,
      "loss": 1.0931,
      "step": 9038
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.763420955201778e-06,
      "loss": 1.2525,
      "step": 9039
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.761940022213995e-06,
      "loss": 1.1183,
      "step": 9040
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.760459089226214e-06,
      "loss": 1.1384,
      "step": 9041
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.75897815623843e-06,
      "loss": 1.3403,
      "step": 9042
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.757497223250649e-06,
      "loss": 1.2024,
      "step": 9043
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.7560162902628654e-06,
      "loss": 1.3207,
      "step": 9044
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.754535357275084e-06,
      "loss": 1.3414,
      "step": 9045
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.753054424287302e-06,
      "loss": 1.254,
      "step": 9046
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.751573491299519e-06,
      "loss": 1.0918,
      "step": 9047
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.750092558311737e-06,
      "loss": 1.1058,
      "step": 9048
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.748611625323954e-06,
      "loss": 1.2155,
      "step": 9049
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.747130692336172e-06,
      "loss": 1.3183,
      "step": 9050
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.745649759348391e-06,
      "loss": 1.191,
      "step": 9051
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.744168826360607e-06,
      "loss": 1.2532,
      "step": 9052
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.742687893372826e-06,
      "loss": 1.3524,
      "step": 9053
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.741206960385043e-06,
      "loss": 1.4087,
      "step": 9054
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.739726027397261e-06,
      "loss": 1.4429,
      "step": 9055
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.738245094409479e-06,
      "loss": 0.996,
      "step": 9056
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.736764161421696e-06,
      "loss": 1.2976,
      "step": 9057
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.735283228433914e-06,
      "loss": 1.1908,
      "step": 9058
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.733802295446132e-06,
      "loss": 1.3719,
      "step": 9059
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.732321362458349e-06,
      "loss": 1.2559,
      "step": 9060
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.730840429470567e-06,
      "loss": 1.3158,
      "step": 9061
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.729359496482785e-06,
      "loss": 1.2149,
      "step": 9062
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.7278785634950025e-06,
      "loss": 1.2993,
      "step": 9063
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.72639763050722e-06,
      "loss": 1.4504,
      "step": 9064
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.724916697519438e-06,
      "loss": 1.1532,
      "step": 9065
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.723435764531655e-06,
      "loss": 1.4659,
      "step": 9066
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.721954831543873e-06,
      "loss": 1.6019,
      "step": 9067
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.720473898556091e-06,
      "loss": 1.4595,
      "step": 9068
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.7189929655683085e-06,
      "loss": 1.4125,
      "step": 9069
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.717512032580527e-06,
      "loss": 1.4994,
      "step": 9070
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.7160310995927435e-06,
      "loss": 1.0905,
      "step": 9071
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.714550166604962e-06,
      "loss": 1.3602,
      "step": 9072
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.71306923361718e-06,
      "loss": 1.2548,
      "step": 9073
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.711588300629397e-06,
      "loss": 1.1547,
      "step": 9074
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.710107367641615e-06,
      "loss": 1.2384,
      "step": 9075
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.708626434653832e-06,
      "loss": 1.2751,
      "step": 9076
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.70714550166605e-06,
      "loss": 1.1548,
      "step": 9077
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.705664568678267e-06,
      "loss": 1.3618,
      "step": 9078
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.704183635690485e-06,
      "loss": 1.2818,
      "step": 9079
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.702702702702704e-06,
      "loss": 1.1667,
      "step": 9080
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.70122176971492e-06,
      "loss": 1.0201,
      "step": 9081
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.699740836727139e-06,
      "loss": 1.2967,
      "step": 9082
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.698259903739356e-06,
      "loss": 1.3158,
      "step": 9083
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.696778970751574e-06,
      "loss": 1.277,
      "step": 9084
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.695298037763792e-06,
      "loss": 1.3561,
      "step": 9085
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.69381710477601e-06,
      "loss": 0.9874,
      "step": 9086
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.692336171788227e-06,
      "loss": 1.2781,
      "step": 9087
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.690855238800445e-06,
      "loss": 1.2803,
      "step": 9088
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.689374305812662e-06,
      "loss": 1.2901,
      "step": 9089
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.6878933728248805e-06,
      "loss": 1.2607,
      "step": 9090
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.686412439837098e-06,
      "loss": 1.1748,
      "step": 9091
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.684931506849316e-06,
      "loss": 1.022,
      "step": 9092
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.683450573861533e-06,
      "loss": 1.2883,
      "step": 9093
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.6819696408737515e-06,
      "loss": 1.4401,
      "step": 9094
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.680488707885969e-06,
      "loss": 1.0489,
      "step": 9095
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.6790077748981865e-06,
      "loss": 1.3712,
      "step": 9096
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.677526841910405e-06,
      "loss": 1.4165,
      "step": 9097
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.6760459089226215e-06,
      "loss": 1.1809,
      "step": 9098
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.67456497593484e-06,
      "loss": 1.2703,
      "step": 9099
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.6730840429470566e-06,
      "loss": 1.5405,
      "step": 9100
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.671603109959275e-06,
      "loss": 1.2762,
      "step": 9101
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.670122176971493e-06,
      "loss": 1.2387,
      "step": 9102
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.66864124398371e-06,
      "loss": 1.0763,
      "step": 9103
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.667160310995928e-06,
      "loss": 1.4607,
      "step": 9104
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.665679378008145e-06,
      "loss": 1.3944,
      "step": 9105
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.664198445020363e-06,
      "loss": 1.1296,
      "step": 9106
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.662717512032582e-06,
      "loss": 1.1083,
      "step": 9107
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.661236579044798e-06,
      "loss": 1.1491,
      "step": 9108
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.659755646057017e-06,
      "loss": 1.32,
      "step": 9109
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.658274713069233e-06,
      "loss": 1.3227,
      "step": 9110
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.656793780081452e-06,
      "loss": 1.2459,
      "step": 9111
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.65531284709367e-06,
      "loss": 1.3123,
      "step": 9112
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.653831914105887e-06,
      "loss": 1.0557,
      "step": 9113
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.652350981118105e-06,
      "loss": 1.3151,
      "step": 9114
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.650870048130323e-06,
      "loss": 1.1888,
      "step": 9115
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.64938911514254e-06,
      "loss": 1.4172,
      "step": 9116
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.647908182154758e-06,
      "loss": 1.4238,
      "step": 9117
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.646427249166976e-06,
      "loss": 1.1707,
      "step": 9118
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.644946316179194e-06,
      "loss": 1.1099,
      "step": 9119
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.643465383191411e-06,
      "loss": 1.1571,
      "step": 9120
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.641984450203629e-06,
      "loss": 1.4082,
      "step": 9121
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.640503517215846e-06,
      "loss": 1.296,
      "step": 9122
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.6390225842280645e-06,
      "loss": 1.1688,
      "step": 9123
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.637541651240282e-06,
      "loss": 1.4077,
      "step": 9124
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.6360607182524996e-06,
      "loss": 1.0937,
      "step": 9125
    },
    {
      "epoch": 2.01,
      "eval_loss": NaN,
      "eval_runtime": 4.3807,
      "eval_samples_per_second": 352.455,
      "eval_steps_per_second": 44.057,
      "step": 9125
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.634579785264718e-06,
      "loss": 1.2877,
      "step": 9126
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.633098852276935e-06,
      "loss": 1.2138,
      "step": 9127
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.631617919289153e-06,
      "loss": 1.372,
      "step": 9128
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.630136986301371e-06,
      "loss": 1.1051,
      "step": 9129
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.628656053313588e-06,
      "loss": 1.3157,
      "step": 9130
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.627175120325806e-06,
      "loss": 1.2794,
      "step": 9131
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.625694187338023e-06,
      "loss": 1.4031,
      "step": 9132
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.624213254350241e-06,
      "loss": 1.3779,
      "step": 9133
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.622732321362458e-06,
      "loss": 1.3,
      "step": 9134
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.6212513883746764e-06,
      "loss": 1.4513,
      "step": 9135
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.619770455386895e-06,
      "loss": 1.2762,
      "step": 9136
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.6182895223991115e-06,
      "loss": 1.2305,
      "step": 9137
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.61680858941133e-06,
      "loss": 1.299,
      "step": 9138
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.6153276564235465e-06,
      "loss": 1.1877,
      "step": 9139
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.613846723435765e-06,
      "loss": 1.2698,
      "step": 9140
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.612365790447983e-06,
      "loss": 1.3561,
      "step": 9141
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.6108848574602e-06,
      "loss": 1.0697,
      "step": 9142
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.609403924472418e-06,
      "loss": 1.1322,
      "step": 9143
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.607922991484636e-06,
      "loss": 1.4078,
      "step": 9144
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.606442058496853e-06,
      "loss": 1.1238,
      "step": 9145
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.604961125509072e-06,
      "loss": 1.1841,
      "step": 9146
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.603480192521289e-06,
      "loss": 1.1483,
      "step": 9147
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.601999259533507e-06,
      "loss": 1.2406,
      "step": 9148
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.600518326545724e-06,
      "loss": 1.2977,
      "step": 9149
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.5990373935579426e-06,
      "loss": 1.1607,
      "step": 9150
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.597556460570159e-06,
      "loss": 1.1801,
      "step": 9151
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.596075527582378e-06,
      "loss": 1.2442,
      "step": 9152
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.594594594594595e-06,
      "loss": 1.3153,
      "step": 9153
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.593113661606813e-06,
      "loss": 1.1177,
      "step": 9154
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.591632728619031e-06,
      "loss": 1.4392,
      "step": 9155
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.590151795631248e-06,
      "loss": 1.2903,
      "step": 9156
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.588670862643466e-06,
      "loss": 1.3539,
      "step": 9157
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.587189929655684e-06,
      "loss": 1.3267,
      "step": 9158
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.585708996667901e-06,
      "loss": 1.2561,
      "step": 9159
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.5842280636801194e-06,
      "loss": 0.9934,
      "step": 9160
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.582747130692336e-06,
      "loss": 1.2794,
      "step": 9161
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.5812661977045545e-06,
      "loss": 1.2238,
      "step": 9162
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.579785264716773e-06,
      "loss": 1.0257,
      "step": 9163
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.5783043317289895e-06,
      "loss": 1.133,
      "step": 9164
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.576823398741208e-06,
      "loss": 1.5735,
      "step": 9165
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.5753424657534245e-06,
      "loss": 1.2339,
      "step": 9166
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.573861532765643e-06,
      "loss": 1.2012,
      "step": 9167
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.572380599777861e-06,
      "loss": 1.0158,
      "step": 9168
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.570899666790078e-06,
      "loss": 1.2279,
      "step": 9169
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.569418733802296e-06,
      "loss": 1.2387,
      "step": 9170
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.567937800814513e-06,
      "loss": 1.4538,
      "step": 9171
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.566456867826731e-06,
      "loss": 1.4342,
      "step": 9172
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.564975934838949e-06,
      "loss": 1.2611,
      "step": 9173
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.563495001851166e-06,
      "loss": 1.007,
      "step": 9174
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.562014068863385e-06,
      "loss": 1.1985,
      "step": 9175
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.560533135875602e-06,
      "loss": 1.2873,
      "step": 9176
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.55905220288782e-06,
      "loss": 1.1257,
      "step": 9177
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.557571269900037e-06,
      "loss": 1.1226,
      "step": 9178
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.556090336912256e-06,
      "loss": 1.2141,
      "step": 9179
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.554609403924473e-06,
      "loss": 1.2213,
      "step": 9180
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.553128470936691e-06,
      "loss": 1.1424,
      "step": 9181
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.551647537948909e-06,
      "loss": 1.2063,
      "step": 9182
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.550166604961126e-06,
      "loss": 1.2041,
      "step": 9183
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.548685671973344e-06,
      "loss": 1.3549,
      "step": 9184
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.547204738985562e-06,
      "loss": 1.269,
      "step": 9185
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.545723805997779e-06,
      "loss": 1.1735,
      "step": 9186
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.5442428730099975e-06,
      "loss": 1.2476,
      "step": 9187
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.542761940022214e-06,
      "loss": 1.2057,
      "step": 9188
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.5412810070344325e-06,
      "loss": 1.1148,
      "step": 9189
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.539800074046649e-06,
      "loss": 1.2187,
      "step": 9190
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.5383191410588675e-06,
      "loss": 1.2195,
      "step": 9191
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.536838208071086e-06,
      "loss": 1.3177,
      "step": 9192
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.5353572750833026e-06,
      "loss": 1.2637,
      "step": 9193
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.533876342095521e-06,
      "loss": 1.195,
      "step": 9194
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.532395409107738e-06,
      "loss": 1.4072,
      "step": 9195
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.530914476119956e-06,
      "loss": 1.3109,
      "step": 9196
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.529433543132174e-06,
      "loss": 1.1679,
      "step": 9197
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.527952610144391e-06,
      "loss": 1.2408,
      "step": 9198
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.526471677156609e-06,
      "loss": 1.3155,
      "step": 9199
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.524990744168827e-06,
      "loss": 1.1984,
      "step": 9200
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.523509811181044e-06,
      "loss": 1.1206,
      "step": 9201
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.522028878193263e-06,
      "loss": 1.0726,
      "step": 9202
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.5205479452054794e-06,
      "loss": 1.3032,
      "step": 9203
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.519067012217698e-06,
      "loss": 1.2484,
      "step": 9204
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.517586079229915e-06,
      "loss": 1.4272,
      "step": 9205
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.516105146242133e-06,
      "loss": 1.2437,
      "step": 9206
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.51462421325435e-06,
      "loss": 1.2263,
      "step": 9207
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.513143280266569e-06,
      "loss": 1.3336,
      "step": 9208
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.511662347278786e-06,
      "loss": 1.3453,
      "step": 9209
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.510181414291004e-06,
      "loss": 1.2403,
      "step": 9210
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.508700481303222e-06,
      "loss": 1.1434,
      "step": 9211
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.507219548315439e-06,
      "loss": 1.3301,
      "step": 9212
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.505738615327657e-06,
      "loss": 1.3684,
      "step": 9213
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.5042576823398755e-06,
      "loss": 1.3138,
      "step": 9214
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.502776749352092e-06,
      "loss": 1.135,
      "step": 9215
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.5012958163643105e-06,
      "loss": 1.1923,
      "step": 9216
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.499814883376527e-06,
      "loss": 1.0764,
      "step": 9217
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.498333950388746e-06,
      "loss": 1.4009,
      "step": 9218
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.496853017400964e-06,
      "loss": 1.3093,
      "step": 9219
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.495372084413181e-06,
      "loss": 1.2504,
      "step": 9220
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.493891151425399e-06,
      "loss": 1.322,
      "step": 9221
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.492410218437616e-06,
      "loss": 1.245,
      "step": 9222
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.490929285449834e-06,
      "loss": 1.1777,
      "step": 9223
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.489448352462051e-06,
      "loss": 1.2819,
      "step": 9224
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.487967419474269e-06,
      "loss": 1.2409,
      "step": 9225
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.486486486486487e-06,
      "loss": 1.1965,
      "step": 9226
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.485005553498704e-06,
      "loss": 1.3337,
      "step": 9227
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.4835246205109224e-06,
      "loss": 1.3522,
      "step": 9228
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.48204368752314e-06,
      "loss": 1.1493,
      "step": 9229
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.4805627545353575e-06,
      "loss": 1.2056,
      "step": 9230
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.479081821547576e-06,
      "loss": 1.3502,
      "step": 9231
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.477600888559793e-06,
      "loss": 1.144,
      "step": 9232
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.476119955572011e-06,
      "loss": 1.0667,
      "step": 9233
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.474639022584228e-06,
      "loss": 1.0797,
      "step": 9234
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.473158089596446e-06,
      "loss": 1.179,
      "step": 9235
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.471677156608664e-06,
      "loss": 1.2426,
      "step": 9236
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.470196223620882e-06,
      "loss": 1.3411,
      "step": 9237
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.468715290633099e-06,
      "loss": 1.3173,
      "step": 9238
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.467234357645317e-06,
      "loss": 1.2517,
      "step": 9239
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.465753424657535e-06,
      "loss": 1.1639,
      "step": 9240
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.464272491669753e-06,
      "loss": 1.2749,
      "step": 9241
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.46279155868197e-06,
      "loss": 1.2111,
      "step": 9242
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.461310625694189e-06,
      "loss": 1.1689,
      "step": 9243
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.459829692706405e-06,
      "loss": 1.1676,
      "step": 9244
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.458348759718624e-06,
      "loss": 1.3032,
      "step": 9245
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.45686782673084e-06,
      "loss": 1.3998,
      "step": 9246
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.455386893743059e-06,
      "loss": 1.1505,
      "step": 9247
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.453905960755277e-06,
      "loss": 1.4345,
      "step": 9248
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.452425027767494e-06,
      "loss": 1.2029,
      "step": 9249
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.450944094779712e-06,
      "loss": 1.0437,
      "step": 9250
    },
    {
      "epoch": 2.04,
      "eval_loss": NaN,
      "eval_runtime": 4.3874,
      "eval_samples_per_second": 351.919,
      "eval_steps_per_second": 43.99,
      "step": 9250
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.449463161791929e-06,
      "loss": 0.9867,
      "step": 9251
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.447982228804147e-06,
      "loss": 1.3359,
      "step": 9252
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.4465012958163654e-06,
      "loss": 1.21,
      "step": 9253
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.445020362828582e-06,
      "loss": 1.219,
      "step": 9254
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.4435394298408005e-06,
      "loss": 1.3558,
      "step": 9255
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.442058496853017e-06,
      "loss": 1.302,
      "step": 9256
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.4405775638652355e-06,
      "loss": 1.164,
      "step": 9257
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.439096630877454e-06,
      "loss": 1.3378,
      "step": 9258
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.4376156978896705e-06,
      "loss": 1.0957,
      "step": 9259
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.436134764901889e-06,
      "loss": 0.9515,
      "step": 9260
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.4346538319141064e-06,
      "loss": 1.0974,
      "step": 9261
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.433172898926324e-06,
      "loss": 1.1093,
      "step": 9262
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.4316919659385415e-06,
      "loss": 1.0508,
      "step": 9263
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.43021103295076e-06,
      "loss": 1.3269,
      "step": 9264
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.428730099962977e-06,
      "loss": 1.3094,
      "step": 9265
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.427249166975195e-06,
      "loss": 1.0605,
      "step": 9266
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.425768233987412e-06,
      "loss": 1.0573,
      "step": 9267
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.42428730099963e-06,
      "loss": 1.1632,
      "step": 9268
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.422806368011848e-06,
      "loss": 1.1692,
      "step": 9269
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.421325435024066e-06,
      "loss": 1.1376,
      "step": 9270
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.419844502036283e-06,
      "loss": 1.2511,
      "step": 9271
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.418363569048502e-06,
      "loss": 1.1927,
      "step": 9272
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.416882636060718e-06,
      "loss": 1.334,
      "step": 9273
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.415401703072937e-06,
      "loss": 1.2197,
      "step": 9274
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.413920770085155e-06,
      "loss": 1.1373,
      "step": 9275
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.412439837097372e-06,
      "loss": 1.4344,
      "step": 9276
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.41095890410959e-06,
      "loss": 1.2129,
      "step": 9277
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.409477971121807e-06,
      "loss": 1.0559,
      "step": 9278
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.407997038134025e-06,
      "loss": 1.1192,
      "step": 9279
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.406516105146242e-06,
      "loss": 1.2192,
      "step": 9280
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.40503517215846e-06,
      "loss": 1.2101,
      "step": 9281
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.4035542391706785e-06,
      "loss": 1.1466,
      "step": 9282
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.402073306182895e-06,
      "loss": 1.0502,
      "step": 9283
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.4005923731951136e-06,
      "loss": 1.2127,
      "step": 9284
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.399111440207331e-06,
      "loss": 0.9682,
      "step": 9285
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.397630507219549e-06,
      "loss": 1.3984,
      "step": 9286
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.396149574231767e-06,
      "loss": 1.074,
      "step": 9287
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.394668641243984e-06,
      "loss": 1.2261,
      "step": 9288
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.393187708256202e-06,
      "loss": 1.4363,
      "step": 9289
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.3917067752684195e-06,
      "loss": 1.2337,
      "step": 9290
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.390225842280637e-06,
      "loss": 0.9504,
      "step": 9291
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.388744909292855e-06,
      "loss": 1.2906,
      "step": 9292
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.387263976305073e-06,
      "loss": 1.2344,
      "step": 9293
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.38578304331729e-06,
      "loss": 1.1878,
      "step": 9294
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.384302110329508e-06,
      "loss": 1.3748,
      "step": 9295
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.382821177341726e-06,
      "loss": 1.0724,
      "step": 9296
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.381340244353944e-06,
      "loss": 1.0381,
      "step": 9297
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.379859311366161e-06,
      "loss": 1.3528,
      "step": 9298
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.378378378378379e-06,
      "loss": 1.2365,
      "step": 9299
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.376897445390596e-06,
      "loss": 1.2595,
      "step": 9300
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.375416512402815e-06,
      "loss": 1.2799,
      "step": 9301
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.373935579415031e-06,
      "loss": 1.2997,
      "step": 9302
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.37245464642725e-06,
      "loss": 1.1603,
      "step": 9303
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.370973713439468e-06,
      "loss": 1.3937,
      "step": 9304
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.369492780451685e-06,
      "loss": 1.2319,
      "step": 9305
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.368011847463903e-06,
      "loss": 1.3165,
      "step": 9306
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.36653091447612e-06,
      "loss": 1.2234,
      "step": 9307
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.365049981488338e-06,
      "loss": 1.2602,
      "step": 9308
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.3635690485005566e-06,
      "loss": 1.2534,
      "step": 9309
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.362088115512773e-06,
      "loss": 1.081,
      "step": 9310
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.360607182524992e-06,
      "loss": 1.1892,
      "step": 9311
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.359126249537208e-06,
      "loss": 1.2859,
      "step": 9312
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.357645316549427e-06,
      "loss": 1.3121,
      "step": 9313
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.356164383561645e-06,
      "loss": 1.1894,
      "step": 9314
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.354683450573862e-06,
      "loss": 1.2835,
      "step": 9315
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.35320251758608e-06,
      "loss": 1.4076,
      "step": 9316
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.3517215845982975e-06,
      "loss": 1.351,
      "step": 9317
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.350240651610515e-06,
      "loss": 1.1857,
      "step": 9318
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.3487597186227326e-06,
      "loss": 1.4061,
      "step": 9319
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.34727878563495e-06,
      "loss": 1.2984,
      "step": 9320
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.3457978526471685e-06,
      "loss": 1.1146,
      "step": 9321
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.344316919659386e-06,
      "loss": 1.5022,
      "step": 9322
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.3428359866716035e-06,
      "loss": 1.1075,
      "step": 9323
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.341355053683821e-06,
      "loss": 1.0893,
      "step": 9324
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.339874120696039e-06,
      "loss": 1.0907,
      "step": 9325
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.338393187708257e-06,
      "loss": 1.3856,
      "step": 9326
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.336912254720474e-06,
      "loss": 1.521,
      "step": 9327
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.335431321732693e-06,
      "loss": 1.0946,
      "step": 9328
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.3339503887449094e-06,
      "loss": 1.1683,
      "step": 9329
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.332469455757128e-06,
      "loss": 1.1008,
      "step": 9330
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.330988522769345e-06,
      "loss": 1.3362,
      "step": 9331
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.329507589781563e-06,
      "loss": 1.2383,
      "step": 9332
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.328026656793781e-06,
      "loss": 1.1232,
      "step": 9333
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.326545723805998e-06,
      "loss": 1.1163,
      "step": 9334
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.325064790818216e-06,
      "loss": 1.1821,
      "step": 9335
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.323583857830433e-06,
      "loss": 1.1304,
      "step": 9336
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.322102924842651e-06,
      "loss": 1.2049,
      "step": 9337
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.32062199185487e-06,
      "loss": 1.0758,
      "step": 9338
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.319141058867086e-06,
      "loss": 1.2605,
      "step": 9339
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.317660125879305e-06,
      "loss": 1.1951,
      "step": 9340
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.316179192891521e-06,
      "loss": 1.3535,
      "step": 9341
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.31469825990374e-06,
      "loss": 1.2815,
      "step": 9342
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.313217326915958e-06,
      "loss": 1.1977,
      "step": 9343
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.311736393928175e-06,
      "loss": 1.2726,
      "step": 9344
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.310255460940393e-06,
      "loss": 1.1085,
      "step": 9345
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.308774527952611e-06,
      "loss": 1.1404,
      "step": 9346
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.307293594964828e-06,
      "loss": 1.2839,
      "step": 9347
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.3058126619770465e-06,
      "loss": 0.9673,
      "step": 9348
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.304331728989264e-06,
      "loss": 1.5256,
      "step": 9349
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.3028507960014815e-06,
      "loss": 1.278,
      "step": 9350
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.301369863013699e-06,
      "loss": 1.1849,
      "step": 9351
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.2998889300259166e-06,
      "loss": 1.0942,
      "step": 9352
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.298407997038134e-06,
      "loss": 1.3521,
      "step": 9353
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.2969270640503524e-06,
      "loss": 0.94,
      "step": 9354
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.29544613106257e-06,
      "loss": 1.2304,
      "step": 9355
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.2939651980747875e-06,
      "loss": 1.213,
      "step": 9356
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.292484265087006e-06,
      "loss": 1.1561,
      "step": 9357
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.2910033320992225e-06,
      "loss": 1.0938,
      "step": 9358
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.289522399111441e-06,
      "loss": 1.0502,
      "step": 9359
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.288041466123659e-06,
      "loss": 1.3594,
      "step": 9360
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.286560533135876e-06,
      "loss": 1.2347,
      "step": 9361
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.285079600148094e-06,
      "loss": 0.9916,
      "step": 9362
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.283598667160311e-06,
      "loss": 1.3273,
      "step": 9363
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.282117734172529e-06,
      "loss": 1.1418,
      "step": 9364
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.280636801184748e-06,
      "loss": 1.3371,
      "step": 9365
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.279155868196964e-06,
      "loss": 1.2103,
      "step": 9366
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.277674935209183e-06,
      "loss": 1.0168,
      "step": 9367
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.276194002221399e-06,
      "loss": 1.1016,
      "step": 9368
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.274713069233618e-06,
      "loss": 1.2601,
      "step": 9369
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.273232136245836e-06,
      "loss": 1.2285,
      "step": 9370
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.271751203258053e-06,
      "loss": 1.1979,
      "step": 9371
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.270270270270271e-06,
      "loss": 1.4206,
      "step": 9372
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.268789337282488e-06,
      "loss": 1.0376,
      "step": 9373
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.267308404294706e-06,
      "loss": 1.4989,
      "step": 9374
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.265827471306924e-06,
      "loss": 1.1533,
      "step": 9375
    },
    {
      "epoch": 2.07,
      "eval_loss": NaN,
      "eval_runtime": 4.3736,
      "eval_samples_per_second": 353.027,
      "eval_steps_per_second": 44.128,
      "step": 9375
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.264346538319141e-06,
      "loss": 1.2181,
      "step": 9376
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.2628656053313596e-06,
      "loss": 1.2593,
      "step": 9377
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.261384672343577e-06,
      "loss": 1.1845,
      "step": 9378
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.259903739355795e-06,
      "loss": 1.2626,
      "step": 9379
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.258422806368012e-06,
      "loss": 1.0656,
      "step": 9380
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.2569418733802305e-06,
      "loss": 1.1432,
      "step": 9381
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.255460940392448e-06,
      "loss": 1.3502,
      "step": 9382
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.2539800074046655e-06,
      "loss": 1.3504,
      "step": 9383
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.252499074416883e-06,
      "loss": 1.4251,
      "step": 9384
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.2510181414291005e-06,
      "loss": 1.1562,
      "step": 9385
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.249537208441319e-06,
      "loss": 1.0903,
      "step": 9386
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.2480562754535364e-06,
      "loss": 1.1777,
      "step": 9387
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.246575342465754e-06,
      "loss": 1.2665,
      "step": 9388
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.245094409477972e-06,
      "loss": 1.0418,
      "step": 9389
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.243613476490189e-06,
      "loss": 1.2835,
      "step": 9390
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.242132543502407e-06,
      "loss": 1.2919,
      "step": 9391
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.240651610514624e-06,
      "loss": 1.2116,
      "step": 9392
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.239170677526842e-06,
      "loss": 1.2668,
      "step": 9393
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.237689744539061e-06,
      "loss": 1.2253,
      "step": 9394
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.236208811551277e-06,
      "loss": 1.1948,
      "step": 9395
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.234727878563496e-06,
      "loss": 1.0512,
      "step": 9396
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.2332469455757124e-06,
      "loss": 1.0874,
      "step": 9397
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.231766012587931e-06,
      "loss": 1.321,
      "step": 9398
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.230285079600149e-06,
      "loss": 1.155,
      "step": 9399
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.228804146612366e-06,
      "loss": 1.1808,
      "step": 9400
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.227323213624584e-06,
      "loss": 1.2463,
      "step": 9401
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.225842280636801e-06,
      "loss": 1.1862,
      "step": 9402
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.224361347649019e-06,
      "loss": 1.1077,
      "step": 9403
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.222880414661238e-06,
      "loss": 1.3053,
      "step": 9404
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.221399481673454e-06,
      "loss": 1.1758,
      "step": 9405
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.219918548685673e-06,
      "loss": 1.3415,
      "step": 9406
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.21843761569789e-06,
      "loss": 1.3399,
      "step": 9407
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.216956682710108e-06,
      "loss": 1.1513,
      "step": 9408
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.215475749722325e-06,
      "loss": 1.2229,
      "step": 9409
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.2139948167345435e-06,
      "loss": 1.2658,
      "step": 9410
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.212513883746761e-06,
      "loss": 1.1299,
      "step": 9411
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.211032950758979e-06,
      "loss": 1.1516,
      "step": 9412
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.209552017771197e-06,
      "loss": 1.0251,
      "step": 9413
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.208071084783414e-06,
      "loss": 1.2229,
      "step": 9414
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.206590151795632e-06,
      "loss": 1.0214,
      "step": 9415
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.2051092188078495e-06,
      "loss": 1.2009,
      "step": 9416
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.203628285820067e-06,
      "loss": 1.0395,
      "step": 9417
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.202147352832285e-06,
      "loss": 1.2059,
      "step": 9418
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.200666419844502e-06,
      "loss": 1.3646,
      "step": 9419
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.19918548685672e-06,
      "loss": 1.1401,
      "step": 9420
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.197704553868939e-06,
      "loss": 1.1496,
      "step": 9421
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.1962236208811554e-06,
      "loss": 1.2479,
      "step": 9422
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.194742687893374e-06,
      "loss": 1.1904,
      "step": 9423
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.1932617549055905e-06,
      "loss": 1.2286,
      "step": 9424
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.191780821917809e-06,
      "loss": 1.0224,
      "step": 9425
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.1902998889300255e-06,
      "loss": 1.2153,
      "step": 9426
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.188818955942244e-06,
      "loss": 1.0524,
      "step": 9427
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.187338022954462e-06,
      "loss": 1.3284,
      "step": 9428
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.185857089966679e-06,
      "loss": 1.1692,
      "step": 9429
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.184376156978897e-06,
      "loss": 1.4141,
      "step": 9430
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.182895223991115e-06,
      "loss": 1.1472,
      "step": 9431
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.181414291003332e-06,
      "loss": 1.2903,
      "step": 9432
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.179933358015551e-06,
      "loss": 1.1843,
      "step": 9433
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.178452425027767e-06,
      "loss": 1.3691,
      "step": 9434
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.176971492039986e-06,
      "loss": 1.3692,
      "step": 9435
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.175490559052203e-06,
      "loss": 1.2064,
      "step": 9436
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.174009626064421e-06,
      "loss": 1.2716,
      "step": 9437
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.172528693076639e-06,
      "loss": 1.2565,
      "step": 9438
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.171047760088857e-06,
      "loss": 1.1367,
      "step": 9439
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.169566827101074e-06,
      "loss": 1.4594,
      "step": 9440
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.168085894113292e-06,
      "loss": 1.0713,
      "step": 9441
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.16660496112551e-06,
      "loss": 1.1441,
      "step": 9442
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.1651240281377275e-06,
      "loss": 1.3544,
      "step": 9443
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.163643095149945e-06,
      "loss": 1.4196,
      "step": 9444
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.162162162162163e-06,
      "loss": 1.2838,
      "step": 9445
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.16068122917438e-06,
      "loss": 1.0909,
      "step": 9446
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.1592002961865985e-06,
      "loss": 1.2586,
      "step": 9447
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.157719363198815e-06,
      "loss": 1.267,
      "step": 9448
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.1562384302110335e-06,
      "loss": 1.2054,
      "step": 9449
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.154757497223252e-06,
      "loss": 1.2042,
      "step": 9450
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.1532765642354685e-06,
      "loss": 1.2076,
      "step": 9451
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.151795631247687e-06,
      "loss": 1.1896,
      "step": 9452
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.1503146982599036e-06,
      "loss": 1.1595,
      "step": 9453
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.148833765272122e-06,
      "loss": 0.9882,
      "step": 9454
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.14735283228434e-06,
      "loss": 1.0934,
      "step": 9455
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.145871899296557e-06,
      "loss": 1.0266,
      "step": 9456
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.144390966308775e-06,
      "loss": 1.229,
      "step": 9457
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.142910033320992e-06,
      "loss": 1.2628,
      "step": 9458
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.14142910033321e-06,
      "loss": 1.5416,
      "step": 9459
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.139948167345429e-06,
      "loss": 1.1881,
      "step": 9460
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.138467234357645e-06,
      "loss": 1.035,
      "step": 9461
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.136986301369864e-06,
      "loss": 1.0278,
      "step": 9462
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.135505368382081e-06,
      "loss": 1.4538,
      "step": 9463
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.134024435394299e-06,
      "loss": 1.1184,
      "step": 9464
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.132543502406516e-06,
      "loss": 1.2515,
      "step": 9465
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.131062569418734e-06,
      "loss": 1.2713,
      "step": 9466
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.129581636430952e-06,
      "loss": 1.0937,
      "step": 9467
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.12810070344317e-06,
      "loss": 1.1993,
      "step": 9468
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.126619770455387e-06,
      "loss": 1.0577,
      "step": 9469
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.125138837467605e-06,
      "loss": 1.1967,
      "step": 9470
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.123657904479823e-06,
      "loss": 1.1584,
      "step": 9471
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.122176971492041e-06,
      "loss": 1.0216,
      "step": 9472
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.120696038504258e-06,
      "loss": 1.3409,
      "step": 9473
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.1192151055164765e-06,
      "loss": 1.1998,
      "step": 9474
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.117734172528693e-06,
      "loss": 1.2974,
      "step": 9475
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.1162532395409115e-06,
      "loss": 1.157,
      "step": 9476
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.11477230655313e-06,
      "loss": 1.1189,
      "step": 9477
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.1132913735653466e-06,
      "loss": 1.2913,
      "step": 9478
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.111810440577565e-06,
      "loss": 1.1653,
      "step": 9479
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.110329507589782e-06,
      "loss": 1.1495,
      "step": 9480
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.108848574602e-06,
      "loss": 1.1057,
      "step": 9481
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.107367641614217e-06,
      "loss": 1.2798,
      "step": 9482
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.105886708626435e-06,
      "loss": 1.2844,
      "step": 9483
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.104405775638653e-06,
      "loss": 1.105,
      "step": 9484
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.10292484265087e-06,
      "loss": 1.188,
      "step": 9485
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.101443909663088e-06,
      "loss": 1.1108,
      "step": 9486
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.099962976675305e-06,
      "loss": 1.0569,
      "step": 9487
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.098482043687523e-06,
      "loss": 1.217,
      "step": 9488
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.097001110699742e-06,
      "loss": 1.267,
      "step": 9489
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.0955201777119585e-06,
      "loss": 1.2704,
      "step": 9490
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.094039244724177e-06,
      "loss": 1.2012,
      "step": 9491
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.092558311736394e-06,
      "loss": 1.2187,
      "step": 9492
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.091077378748612e-06,
      "loss": 1.3519,
      "step": 9493
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.08959644576083e-06,
      "loss": 1.2217,
      "step": 9494
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.088115512773048e-06,
      "loss": 1.2545,
      "step": 9495
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.086634579785265e-06,
      "loss": 1.0705,
      "step": 9496
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.085153646797483e-06,
      "loss": 1.1194,
      "step": 9497
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.0836727138097e-06,
      "loss": 1.2858,
      "step": 9498
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.082191780821919e-06,
      "loss": 1.1862,
      "step": 9499
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.080710847834136e-06,
      "loss": 1.4161,
      "step": 9500
    },
    {
      "epoch": 2.09,
      "eval_loss": NaN,
      "eval_runtime": 4.3782,
      "eval_samples_per_second": 352.66,
      "eval_steps_per_second": 44.082,
      "step": 9500
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.079229914846354e-06,
      "loss": 1.2263,
      "step": 9501
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.077748981858571e-06,
      "loss": 1.1822,
      "step": 9502
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0762680488707896e-06,
      "loss": 1.0869,
      "step": 9503
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.074787115883006e-06,
      "loss": 1.3588,
      "step": 9504
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.073306182895225e-06,
      "loss": 1.0595,
      "step": 9505
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.071825249907443e-06,
      "loss": 1.185,
      "step": 9506
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.07034431691966e-06,
      "loss": 1.1289,
      "step": 9507
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.068863383931878e-06,
      "loss": 1.1614,
      "step": 9508
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.067382450944095e-06,
      "loss": 1.2744,
      "step": 9509
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.065901517956313e-06,
      "loss": 1.4275,
      "step": 9510
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.064420584968531e-06,
      "loss": 1.3796,
      "step": 9511
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.062939651980748e-06,
      "loss": 1.2906,
      "step": 9512
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0614587189929664e-06,
      "loss": 1.233,
      "step": 9513
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.059977786005183e-06,
      "loss": 1.1297,
      "step": 9514
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0584968530174015e-06,
      "loss": 1.2443,
      "step": 9515
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.05701592002962e-06,
      "loss": 1.2768,
      "step": 9516
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0555349870418365e-06,
      "loss": 1.0871,
      "step": 9517
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.054054054054055e-06,
      "loss": 1.2125,
      "step": 9518
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0525731210662715e-06,
      "loss": 1.2186,
      "step": 9519
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.05109218807849e-06,
      "loss": 0.9308,
      "step": 9520
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.049611255090707e-06,
      "loss": 1.1326,
      "step": 9521
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.048130322102925e-06,
      "loss": 1.018,
      "step": 9522
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.046649389115143e-06,
      "loss": 1.2244,
      "step": 9523
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.045168456127361e-06,
      "loss": 1.3763,
      "step": 9524
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.043687523139578e-06,
      "loss": 1.0886,
      "step": 9525
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.042206590151796e-06,
      "loss": 1.2572,
      "step": 9526
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.040725657164014e-06,
      "loss": 1.3883,
      "step": 9527
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.039244724176232e-06,
      "loss": 1.3681,
      "step": 9528
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.037763791188449e-06,
      "loss": 1.2409,
      "step": 9529
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.036282858200667e-06,
      "loss": 1.2213,
      "step": 9530
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.034801925212884e-06,
      "loss": 1.2649,
      "step": 9531
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.033320992225103e-06,
      "loss": 1.0165,
      "step": 9532
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.03184005923732e-06,
      "loss": 1.1379,
      "step": 9533
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.030359126249538e-06,
      "loss": 1.214,
      "step": 9534
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.028878193261756e-06,
      "loss": 1.0209,
      "step": 9535
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.027397260273973e-06,
      "loss": 1.0082,
      "step": 9536
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.025916327286191e-06,
      "loss": 1.2091,
      "step": 9537
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.024435394298408e-06,
      "loss": 1.155,
      "step": 9538
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.022954461310626e-06,
      "loss": 1.4257,
      "step": 9539
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0214735283228445e-06,
      "loss": 1.2137,
      "step": 9540
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.019992595335061e-06,
      "loss": 1.5054,
      "step": 9541
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0185116623472795e-06,
      "loss": 1.3474,
      "step": 9542
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.017030729359496e-06,
      "loss": 1.3246,
      "step": 9543
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0155497963717145e-06,
      "loss": 1.1119,
      "step": 9544
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.014068863383933e-06,
      "loss": 0.9816,
      "step": 9545
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0125879303961496e-06,
      "loss": 1.3053,
      "step": 9546
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.011106997408368e-06,
      "loss": 1.2665,
      "step": 9547
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.0096260644205854e-06,
      "loss": 1.4263,
      "step": 9548
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.008145131432803e-06,
      "loss": 1.1396,
      "step": 9549
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.006664198445021e-06,
      "loss": 1.3063,
      "step": 9550
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.005183265457238e-06,
      "loss": 1.1988,
      "step": 9551
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.003702332469456e-06,
      "loss": 1.2397,
      "step": 9552
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.002221399481674e-06,
      "loss": 1.2352,
      "step": 9553
    },
    {
      "epoch": 2.11,
      "learning_rate": 6.000740466493891e-06,
      "loss": 1.0361,
      "step": 9554
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.999259533506109e-06,
      "loss": 1.227,
      "step": 9555
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.997778600518327e-06,
      "loss": 1.1272,
      "step": 9556
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.996297667530545e-06,
      "loss": 1.1437,
      "step": 9557
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.994816734542762e-06,
      "loss": 1.4373,
      "step": 9558
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.993335801554981e-06,
      "loss": 1.0637,
      "step": 9559
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.991854868567197e-06,
      "loss": 1.1851,
      "step": 9560
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.990373935579416e-06,
      "loss": 1.4609,
      "step": 9561
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.988893002591633e-06,
      "loss": 1.1549,
      "step": 9562
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.987412069603851e-06,
      "loss": 1.303,
      "step": 9563
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.985931136616069e-06,
      "loss": 1.1745,
      "step": 9564
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.984450203628286e-06,
      "loss": 1.4236,
      "step": 9565
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.982969270640504e-06,
      "loss": 1.1404,
      "step": 9566
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.9814883376527225e-06,
      "loss": 1.302,
      "step": 9567
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.980007404664939e-06,
      "loss": 1.204,
      "step": 9568
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.9785264716771575e-06,
      "loss": 1.0488,
      "step": 9569
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.977045538689374e-06,
      "loss": 1.6144,
      "step": 9570
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.9755646057015926e-06,
      "loss": 1.2611,
      "step": 9571
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.974083672713811e-06,
      "loss": 0.946,
      "step": 9572
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.972602739726028e-06,
      "loss": 1.2282,
      "step": 9573
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.971121806738246e-06,
      "loss": 1.3856,
      "step": 9574
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.969640873750463e-06,
      "loss": 1.3986,
      "step": 9575
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.968159940762681e-06,
      "loss": 1.1284,
      "step": 9576
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.9666790077748985e-06,
      "loss": 1.199,
      "step": 9577
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.965198074787116e-06,
      "loss": 1.2165,
      "step": 9578
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.963717141799334e-06,
      "loss": 1.2512,
      "step": 9579
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.962236208811552e-06,
      "loss": 1.1114,
      "step": 9580
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.9607552758237694e-06,
      "loss": 1.1654,
      "step": 9581
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.959274342835987e-06,
      "loss": 1.1123,
      "step": 9582
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.9577934098482045e-06,
      "loss": 1.3245,
      "step": 9583
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.956312476860423e-06,
      "loss": 1.4456,
      "step": 9584
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.95483154387264e-06,
      "loss": 1.1019,
      "step": 9585
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.953350610884858e-06,
      "loss": 1.0973,
      "step": 9586
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.951869677897075e-06,
      "loss": 1.4411,
      "step": 9587
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.950388744909294e-06,
      "loss": 1.4224,
      "step": 9588
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.948907811921511e-06,
      "loss": 1.2551,
      "step": 9589
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.947426878933729e-06,
      "loss": 1.2,
      "step": 9590
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.945945945945947e-06,
      "loss": 1.2044,
      "step": 9591
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.944465012958164e-06,
      "loss": 1.1891,
      "step": 9592
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.942984079970382e-06,
      "loss": 1.3366,
      "step": 9593
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.941503146982599e-06,
      "loss": 1.2911,
      "step": 9594
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.940022213994817e-06,
      "loss": 1.0402,
      "step": 9595
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.9385412810070356e-06,
      "loss": 1.1207,
      "step": 9596
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.937060348019252e-06,
      "loss": 1.0167,
      "step": 9597
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.935579415031471e-06,
      "loss": 1.0398,
      "step": 9598
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.934098482043687e-06,
      "loss": 1.2232,
      "step": 9599
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.932617549055906e-06,
      "loss": 1.3187,
      "step": 9600
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.931136616068124e-06,
      "loss": 1.3327,
      "step": 9601
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.929655683080341e-06,
      "loss": 1.2651,
      "step": 9602
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.928174750092559e-06,
      "loss": 1.0714,
      "step": 9603
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.926693817104776e-06,
      "loss": 1.1997,
      "step": 9604
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.925212884116994e-06,
      "loss": 1.5443,
      "step": 9605
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.9237319511292124e-06,
      "loss": 1.0723,
      "step": 9606
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.922251018141429e-06,
      "loss": 1.3897,
      "step": 9607
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.9207700851536475e-06,
      "loss": 1.1211,
      "step": 9608
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.919289152165865e-06,
      "loss": 1.3538,
      "step": 9609
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.9178082191780825e-06,
      "loss": 1.161,
      "step": 9610
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.9163272861903e-06,
      "loss": 1.1769,
      "step": 9611
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.914846353202518e-06,
      "loss": 1.2893,
      "step": 9612
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.913365420214736e-06,
      "loss": 1.1504,
      "step": 9613
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.911884487226953e-06,
      "loss": 1.0824,
      "step": 9614
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.910403554239171e-06,
      "loss": 1.4846,
      "step": 9615
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.9089226212513885e-06,
      "loss": 1.1276,
      "step": 9616
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.907441688263607e-06,
      "loss": 1.339,
      "step": 9617
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.905960755275824e-06,
      "loss": 1.1618,
      "step": 9618
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.904479822288042e-06,
      "loss": 1.2414,
      "step": 9619
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.90299888930026e-06,
      "loss": 1.3128,
      "step": 9620
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.901517956312477e-06,
      "loss": 1.1988,
      "step": 9621
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.900037023324695e-06,
      "loss": 1.1823,
      "step": 9622
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.898556090336914e-06,
      "loss": 1.3895,
      "step": 9623
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.89707515734913e-06,
      "loss": 1.2728,
      "step": 9624
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.895594224361349e-06,
      "loss": 1.2733,
      "step": 9625
    },
    {
      "epoch": 2.12,
      "eval_loss": NaN,
      "eval_runtime": 4.379,
      "eval_samples_per_second": 352.59,
      "eval_steps_per_second": 44.074,
      "step": 9625
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.894113291373565e-06,
      "loss": 1.0442,
      "step": 9626
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.892632358385784e-06,
      "loss": 1.1931,
      "step": 9627
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.891151425398e-06,
      "loss": 1.2363,
      "step": 9628
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.889670492410219e-06,
      "loss": 1.2641,
      "step": 9629
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.888189559422437e-06,
      "loss": 1.1344,
      "step": 9630
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.886708626434654e-06,
      "loss": 1.1835,
      "step": 9631
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.885227693446872e-06,
      "loss": 1.1697,
      "step": 9632
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.88374676045909e-06,
      "loss": 1.0651,
      "step": 9633
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.882265827471307e-06,
      "loss": 1.2012,
      "step": 9634
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.8807848944835255e-06,
      "loss": 1.2714,
      "step": 9635
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.879303961495742e-06,
      "loss": 1.2871,
      "step": 9636
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8778230285079605e-06,
      "loss": 1.202,
      "step": 9637
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.876342095520178e-06,
      "loss": 1.2418,
      "step": 9638
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8748611625323956e-06,
      "loss": 1.1668,
      "step": 9639
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.873380229544614e-06,
      "loss": 0.9698,
      "step": 9640
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8718992965568315e-06,
      "loss": 1.254,
      "step": 9641
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.870418363569049e-06,
      "loss": 1.2269,
      "step": 9642
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8689374305812665e-06,
      "loss": 1.0779,
      "step": 9643
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.867456497593485e-06,
      "loss": 1.3065,
      "step": 9644
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.865975564605702e-06,
      "loss": 1.1979,
      "step": 9645
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.86449463161792e-06,
      "loss": 1.1584,
      "step": 9646
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.863013698630137e-06,
      "loss": 1.4661,
      "step": 9647
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.861532765642355e-06,
      "loss": 1.0554,
      "step": 9648
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.860051832654573e-06,
      "loss": 1.2652,
      "step": 9649
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.85857089966679e-06,
      "loss": 1.5186,
      "step": 9650
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.857089966679008e-06,
      "loss": 0.966,
      "step": 9651
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.855609033691227e-06,
      "loss": 1.154,
      "step": 9652
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.854128100703443e-06,
      "loss": 1.2006,
      "step": 9653
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.852647167715662e-06,
      "loss": 1.3019,
      "step": 9654
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.851166234727878e-06,
      "loss": 1.1178,
      "step": 9655
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.849685301740097e-06,
      "loss": 1.1396,
      "step": 9656
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.848204368752315e-06,
      "loss": 1.1754,
      "step": 9657
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.846723435764532e-06,
      "loss": 1.1525,
      "step": 9658
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.84524250277675e-06,
      "loss": 1.1177,
      "step": 9659
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.843761569788967e-06,
      "loss": 1.3098,
      "step": 9660
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.842280636801185e-06,
      "loss": 1.109,
      "step": 9661
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8407997038134035e-06,
      "loss": 1.1833,
      "step": 9662
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.83931877082562e-06,
      "loss": 1.2783,
      "step": 9663
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.837837837837839e-06,
      "loss": 1.0207,
      "step": 9664
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.836356904850056e-06,
      "loss": 1.0712,
      "step": 9665
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.834875971862274e-06,
      "loss": 1.2362,
      "step": 9666
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.833395038874491e-06,
      "loss": 1.289,
      "step": 9667
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.831914105886709e-06,
      "loss": 1.138,
      "step": 9668
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.830433172898927e-06,
      "loss": 1.4157,
      "step": 9669
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8289522399111445e-06,
      "loss": 1.371,
      "step": 9670
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.827471306923362e-06,
      "loss": 1.1327,
      "step": 9671
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8259903739355796e-06,
      "loss": 1.0482,
      "step": 9672
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.824509440947798e-06,
      "loss": 1.3548,
      "step": 9673
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8230285079600154e-06,
      "loss": 1.31,
      "step": 9674
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.821547574972233e-06,
      "loss": 1.2678,
      "step": 9675
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.820066641984451e-06,
      "loss": 1.1613,
      "step": 9676
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.818585708996668e-06,
      "loss": 1.17,
      "step": 9677
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.817104776008886e-06,
      "loss": 1.2925,
      "step": 9678
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.815623843021104e-06,
      "loss": 0.9318,
      "step": 9679
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.814142910033321e-06,
      "loss": 1.0461,
      "step": 9680
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.81266197704554e-06,
      "loss": 1.1713,
      "step": 9681
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.8111810440577564e-06,
      "loss": 1.1347,
      "step": 9682
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.809700111069975e-06,
      "loss": 1.1415,
      "step": 9683
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.8082191780821915e-06,
      "loss": 1.1964,
      "step": 9684
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.80673824509441e-06,
      "loss": 1.2232,
      "step": 9685
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.805257312106628e-06,
      "loss": 1.4281,
      "step": 9686
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.803776379118845e-06,
      "loss": 1.222,
      "step": 9687
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.802295446131063e-06,
      "loss": 1.0579,
      "step": 9688
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.80081451314328e-06,
      "loss": 1.0444,
      "step": 9689
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.799333580155498e-06,
      "loss": 1.2153,
      "step": 9690
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.797852647167717e-06,
      "loss": 1.3456,
      "step": 9691
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.796371714179933e-06,
      "loss": 1.2347,
      "step": 9692
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.794890781192152e-06,
      "loss": 1.2227,
      "step": 9693
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.793409848204369e-06,
      "loss": 1.3094,
      "step": 9694
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.791928915216587e-06,
      "loss": 1.2141,
      "step": 9695
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.790447982228805e-06,
      "loss": 1.2949,
      "step": 9696
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.7889670492410226e-06,
      "loss": 1.1203,
      "step": 9697
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.78748611625324e-06,
      "loss": 1.1327,
      "step": 9698
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.786005183265458e-06,
      "loss": 1.2783,
      "step": 9699
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.784524250277675e-06,
      "loss": 1.648,
      "step": 9700
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.7830433172898935e-06,
      "loss": 1.0568,
      "step": 9701
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.781562384302111e-06,
      "loss": 1.0828,
      "step": 9702
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.7800814513143285e-06,
      "loss": 1.203,
      "step": 9703
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.778600518326546e-06,
      "loss": 1.1952,
      "step": 9704
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.777119585338764e-06,
      "loss": 1.1527,
      "step": 9705
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.775638652350981e-06,
      "loss": 1.1977,
      "step": 9706
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.7741577193631994e-06,
      "loss": 1.081,
      "step": 9707
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.772676786375418e-06,
      "loss": 1.0754,
      "step": 9708
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.7711958533876345e-06,
      "loss": 1.2827,
      "step": 9709
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.769714920399853e-06,
      "loss": 1.1077,
      "step": 9710
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.7682339874120695e-06,
      "loss": 1.4795,
      "step": 9711
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.766753054424288e-06,
      "loss": 1.3454,
      "step": 9712
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.765272121436506e-06,
      "loss": 1.2115,
      "step": 9713
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.763791188448723e-06,
      "loss": 1.0872,
      "step": 9714
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.762310255460941e-06,
      "loss": 1.0993,
      "step": 9715
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.760829322473158e-06,
      "loss": 1.2854,
      "step": 9716
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.759348389485376e-06,
      "loss": 1.3962,
      "step": 9717
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.757867456497595e-06,
      "loss": 1.2262,
      "step": 9718
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.756386523509811e-06,
      "loss": 1.1792,
      "step": 9719
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.75490559052203e-06,
      "loss": 1.1082,
      "step": 9720
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.753424657534246e-06,
      "loss": 1.1543,
      "step": 9721
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.751943724546465e-06,
      "loss": 1.1855,
      "step": 9722
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.750462791558682e-06,
      "loss": 1.1959,
      "step": 9723
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.7489818585709e-06,
      "loss": 1.4698,
      "step": 9724
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.747500925583118e-06,
      "loss": 1.29,
      "step": 9725
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.746019992595336e-06,
      "loss": 1.3936,
      "step": 9726
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.744539059607553e-06,
      "loss": 1.0917,
      "step": 9727
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.743058126619771e-06,
      "loss": 1.1707,
      "step": 9728
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.741577193631989e-06,
      "loss": 1.2886,
      "step": 9729
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.7400962606442066e-06,
      "loss": 1.1391,
      "step": 9730
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.738615327656424e-06,
      "loss": 1.1057,
      "step": 9731
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.737134394668642e-06,
      "loss": 1.4638,
      "step": 9732
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.735653461680859e-06,
      "loss": 1.3322,
      "step": 9733
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.7341725286930775e-06,
      "loss": 1.1073,
      "step": 9734
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.732691595705295e-06,
      "loss": 1.0068,
      "step": 9735
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.7312106627175125e-06,
      "loss": 1.1271,
      "step": 9736
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.729729729729731e-06,
      "loss": 1.2121,
      "step": 9737
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.7282487967419475e-06,
      "loss": 1.235,
      "step": 9738
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.726767863754166e-06,
      "loss": 1.4449,
      "step": 9739
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.7252869307663826e-06,
      "loss": 1.1097,
      "step": 9740
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.723805997778601e-06,
      "loss": 1.231,
      "step": 9741
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.722325064790819e-06,
      "loss": 1.0258,
      "step": 9742
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.720844131803036e-06,
      "loss": 1.299,
      "step": 9743
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.719363198815254e-06,
      "loss": 1.0849,
      "step": 9744
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.717882265827471e-06,
      "loss": 1.181,
      "step": 9745
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.716401332839689e-06,
      "loss": 1.2275,
      "step": 9746
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.714920399851908e-06,
      "loss": 1.2465,
      "step": 9747
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.713439466864124e-06,
      "loss": 1.15,
      "step": 9748
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.711958533876343e-06,
      "loss": 1.1412,
      "step": 9749
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.7104776008885594e-06,
      "loss": 1.2394,
      "step": 9750
    },
    {
      "epoch": 2.15,
      "eval_loss": NaN,
      "eval_runtime": 4.3758,
      "eval_samples_per_second": 352.847,
      "eval_steps_per_second": 44.106,
      "step": 9750
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.708996667900778e-06,
      "loss": 1.1049,
      "step": 9751
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.707515734912996e-06,
      "loss": 1.3108,
      "step": 9752
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.706034801925213e-06,
      "loss": 1.1725,
      "step": 9753
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.704553868937431e-06,
      "loss": 1.2371,
      "step": 9754
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.703072935949649e-06,
      "loss": 1.3279,
      "step": 9755
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.701592002961866e-06,
      "loss": 1.264,
      "step": 9756
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.700111069974084e-06,
      "loss": 1.2799,
      "step": 9757
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.698630136986302e-06,
      "loss": 1.1996,
      "step": 9758
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.69714920399852e-06,
      "loss": 1.4212,
      "step": 9759
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.695668271010737e-06,
      "loss": 1.1762,
      "step": 9760
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.6941873380229555e-06,
      "loss": 1.5124,
      "step": 9761
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.692706405035172e-06,
      "loss": 1.1629,
      "step": 9762
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.6912254720473905e-06,
      "loss": 1.2184,
      "step": 9763
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.689744539059608e-06,
      "loss": 1.3981,
      "step": 9764
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.6882636060718256e-06,
      "loss": 1.3929,
      "step": 9765
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.686782673084044e-06,
      "loss": 1.285,
      "step": 9766
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.685301740096261e-06,
      "loss": 1.3088,
      "step": 9767
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.683820807108479e-06,
      "loss": 1.3287,
      "step": 9768
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.682339874120697e-06,
      "loss": 1.246,
      "step": 9769
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.680858941132914e-06,
      "loss": 1.3332,
      "step": 9770
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.679378008145132e-06,
      "loss": 1.2088,
      "step": 9771
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.677897075157349e-06,
      "loss": 1.3541,
      "step": 9772
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.676416142169567e-06,
      "loss": 1.2398,
      "step": 9773
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.674935209181786e-06,
      "loss": 1.1029,
      "step": 9774
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.6734542761940024e-06,
      "loss": 1.3187,
      "step": 9775
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.671973343206221e-06,
      "loss": 1.378,
      "step": 9776
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.6704924102184375e-06,
      "loss": 1.2556,
      "step": 9777
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.669011477230656e-06,
      "loss": 1.2439,
      "step": 9778
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.667530544242873e-06,
      "loss": 0.9613,
      "step": 9779
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.666049611255091e-06,
      "loss": 1.3027,
      "step": 9780
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.664568678267309e-06,
      "loss": 1.1967,
      "step": 9781
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.663087745279526e-06,
      "loss": 1.2352,
      "step": 9782
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.661606812291744e-06,
      "loss": 1.1863,
      "step": 9783
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.660125879303962e-06,
      "loss": 1.2058,
      "step": 9784
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.658644946316179e-06,
      "loss": 1.4638,
      "step": 9785
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.657164013328398e-06,
      "loss": 1.1606,
      "step": 9786
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.655683080340615e-06,
      "loss": 1.1913,
      "step": 9787
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.654202147352833e-06,
      "loss": 1.1595,
      "step": 9788
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.65272121436505e-06,
      "loss": 1.1608,
      "step": 9789
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.651240281377269e-06,
      "loss": 1.1768,
      "step": 9790
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.649759348389486e-06,
      "loss": 1.0502,
      "step": 9791
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.648278415401704e-06,
      "loss": 1.3897,
      "step": 9792
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.646797482413922e-06,
      "loss": 1.2326,
      "step": 9793
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.645316549426139e-06,
      "loss": 1.2185,
      "step": 9794
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.643835616438357e-06,
      "loss": 1.2571,
      "step": 9795
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.642354683450574e-06,
      "loss": 1.434,
      "step": 9796
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.640873750462792e-06,
      "loss": 1.2934,
      "step": 9797
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.63939281747501e-06,
      "loss": 1.1719,
      "step": 9798
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.637911884487227e-06,
      "loss": 1.1606,
      "step": 9799
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.6364309514994454e-06,
      "loss": 1.0703,
      "step": 9800
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.634950018511662e-06,
      "loss": 1.1679,
      "step": 9801
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.6334690855238805e-06,
      "loss": 1.0917,
      "step": 9802
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.631988152536099e-06,
      "loss": 1.2465,
      "step": 9803
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.6305072195483155e-06,
      "loss": 1.3976,
      "step": 9804
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.629026286560534e-06,
      "loss": 1.0837,
      "step": 9805
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.6275453535727505e-06,
      "loss": 1.3456,
      "step": 9806
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.626064420584969e-06,
      "loss": 1.2285,
      "step": 9807
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.624583487597187e-06,
      "loss": 0.9434,
      "step": 9808
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.623102554609404e-06,
      "loss": 1.3858,
      "step": 9809
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.621621621621622e-06,
      "loss": 1.2607,
      "step": 9810
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.62014068863384e-06,
      "loss": 1.0476,
      "step": 9811
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.618659755646057e-06,
      "loss": 0.9927,
      "step": 9812
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.617178822658275e-06,
      "loss": 1.2412,
      "step": 9813
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.615697889670492e-06,
      "loss": 1.2075,
      "step": 9814
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.614216956682711e-06,
      "loss": 1.3635,
      "step": 9815
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.612736023694928e-06,
      "loss": 1.0998,
      "step": 9816
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.611255090707146e-06,
      "loss": 1.4105,
      "step": 9817
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.609774157719363e-06,
      "loss": 1.2235,
      "step": 9818
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.608293224731582e-06,
      "loss": 1.0664,
      "step": 9819
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.606812291743799e-06,
      "loss": 1.1959,
      "step": 9820
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.605331358756017e-06,
      "loss": 1.2227,
      "step": 9821
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.603850425768235e-06,
      "loss": 1.1493,
      "step": 9822
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.602369492780452e-06,
      "loss": 1.1651,
      "step": 9823
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.60088855979267e-06,
      "loss": 0.9127,
      "step": 9824
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.5994076268048884e-06,
      "loss": 1.4657,
      "step": 9825
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.597926693817105e-06,
      "loss": 1.2069,
      "step": 9826
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.5964457608293235e-06,
      "loss": 1.395,
      "step": 9827
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.59496482784154e-06,
      "loss": 1.1419,
      "step": 9828
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.5934838948537585e-06,
      "loss": 1.3011,
      "step": 9829
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.592002961865977e-06,
      "loss": 1.2291,
      "step": 9830
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.5905220288781935e-06,
      "loss": 1.1381,
      "step": 9831
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.589041095890412e-06,
      "loss": 1.1577,
      "step": 9832
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.587560162902629e-06,
      "loss": 1.2991,
      "step": 9833
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.586079229914847e-06,
      "loss": 1.285,
      "step": 9834
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.584598296927064e-06,
      "loss": 1.1404,
      "step": 9835
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.583117363939282e-06,
      "loss": 1.2358,
      "step": 9836
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.5816364309515e-06,
      "loss": 1.066,
      "step": 9837
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.580155497963717e-06,
      "loss": 1.1418,
      "step": 9838
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.578674564975935e-06,
      "loss": 1.2859,
      "step": 9839
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.577193631988153e-06,
      "loss": 1.1299,
      "step": 9840
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.57571269900037e-06,
      "loss": 1.1101,
      "step": 9841
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.574231766012589e-06,
      "loss": 1.108,
      "step": 9842
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.572750833024806e-06,
      "loss": 1.3109,
      "step": 9843
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.571269900037024e-06,
      "loss": 1.2294,
      "step": 9844
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.569788967049241e-06,
      "loss": 1.1729,
      "step": 9845
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.568308034061459e-06,
      "loss": 1.1738,
      "step": 9846
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.566827101073677e-06,
      "loss": 1.2367,
      "step": 9847
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.565346168085895e-06,
      "loss": 1.1091,
      "step": 9848
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.563865235098112e-06,
      "loss": 1.3807,
      "step": 9849
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.56238430211033e-06,
      "loss": 1.3271,
      "step": 9850
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.560903369122548e-06,
      "loss": 1.5429,
      "step": 9851
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.559422436134765e-06,
      "loss": 1.3483,
      "step": 9852
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.557941503146983e-06,
      "loss": 1.219,
      "step": 9853
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.5564605701592015e-06,
      "loss": 1.0512,
      "step": 9854
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.554979637171418e-06,
      "loss": 1.051,
      "step": 9855
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.5534987041836366e-06,
      "loss": 1.1581,
      "step": 9856
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.552017771195853e-06,
      "loss": 1.2998,
      "step": 9857
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.550536838208072e-06,
      "loss": 1.0293,
      "step": 9858
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.54905590522029e-06,
      "loss": 1.1692,
      "step": 9859
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.547574972232507e-06,
      "loss": 1.0686,
      "step": 9860
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.546094039244725e-06,
      "loss": 1.1091,
      "step": 9861
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.544613106256942e-06,
      "loss": 1.3475,
      "step": 9862
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.54313217326916e-06,
      "loss": 1.1113,
      "step": 9863
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.541651240281378e-06,
      "loss": 0.9109,
      "step": 9864
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.540170307293595e-06,
      "loss": 1.2327,
      "step": 9865
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.538689374305813e-06,
      "loss": 1.2686,
      "step": 9866
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.53720844131803e-06,
      "loss": 1.2259,
      "step": 9867
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.5357275083302485e-06,
      "loss": 1.1934,
      "step": 9868
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.534246575342466e-06,
      "loss": 1.1576,
      "step": 9869
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.5327656423546835e-06,
      "loss": 1.391,
      "step": 9870
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.531284709366902e-06,
      "loss": 1.2242,
      "step": 9871
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.529803776379119e-06,
      "loss": 1.1828,
      "step": 9872
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.528322843391337e-06,
      "loss": 1.1287,
      "step": 9873
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.526841910403554e-06,
      "loss": 1.172,
      "step": 9874
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.525360977415773e-06,
      "loss": 1.2682,
      "step": 9875
    },
    {
      "epoch": 2.18,
      "eval_loss": NaN,
      "eval_runtime": 4.3719,
      "eval_samples_per_second": 353.161,
      "eval_steps_per_second": 44.145,
      "step": 9875
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.52388004442799e-06,
      "loss": 1.1272,
      "step": 9876
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.522399111440208e-06,
      "loss": 1.3595,
      "step": 9877
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.520918178452425e-06,
      "loss": 1.5085,
      "step": 9878
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.519437245464643e-06,
      "loss": 1.0116,
      "step": 9879
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.517956312476861e-06,
      "loss": 1.0981,
      "step": 9880
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.516475379489079e-06,
      "loss": 1.2817,
      "step": 9881
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.514994446501296e-06,
      "loss": 1.1069,
      "step": 9882
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.513513513513515e-06,
      "loss": 1.3291,
      "step": 9883
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.512032580525731e-06,
      "loss": 1.1074,
      "step": 9884
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.51055164753795e-06,
      "loss": 1.38,
      "step": 9885
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.509070714550166e-06,
      "loss": 1.0262,
      "step": 9886
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.507589781562385e-06,
      "loss": 1.2194,
      "step": 9887
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.506108848574603e-06,
      "loss": 1.1292,
      "step": 9888
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.50462791558682e-06,
      "loss": 1.3461,
      "step": 9889
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.503146982599038e-06,
      "loss": 1.1268,
      "step": 9890
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.501666049611255e-06,
      "loss": 1.3101,
      "step": 9891
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.500185116623473e-06,
      "loss": 1.3356,
      "step": 9892
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.4987041836356915e-06,
      "loss": 1.2793,
      "step": 9893
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.497223250647908e-06,
      "loss": 1.3389,
      "step": 9894
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.4957423176601265e-06,
      "loss": 1.3833,
      "step": 9895
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.494261384672344e-06,
      "loss": 1.2123,
      "step": 9896
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.4927804516845615e-06,
      "loss": 1.0266,
      "step": 9897
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.49129951869678e-06,
      "loss": 1.3106,
      "step": 9898
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.4898185857089966e-06,
      "loss": 1.1964,
      "step": 9899
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.488337652721215e-06,
      "loss": 1.1237,
      "step": 9900
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.4868567197334324e-06,
      "loss": 1.2618,
      "step": 9901
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.48537578674565e-06,
      "loss": 1.2398,
      "step": 9902
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.483894853757868e-06,
      "loss": 1.2128,
      "step": 9903
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.482413920770086e-06,
      "loss": 1.1838,
      "step": 9904
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.480932987782303e-06,
      "loss": 1.3629,
      "step": 9905
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.479452054794521e-06,
      "loss": 1.2327,
      "step": 9906
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.477971121806739e-06,
      "loss": 1.2694,
      "step": 9907
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.476490188818956e-06,
      "loss": 1.2813,
      "step": 9908
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.475009255831174e-06,
      "loss": 1.2403,
      "step": 9909
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.473528322843392e-06,
      "loss": 1.348,
      "step": 9910
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.472047389855609e-06,
      "loss": 1.1243,
      "step": 9911
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.470566456867828e-06,
      "loss": 1.1123,
      "step": 9912
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.469085523880044e-06,
      "loss": 1.0501,
      "step": 9913
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.467604590892263e-06,
      "loss": 1.0408,
      "step": 9914
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.466123657904481e-06,
      "loss": 1.1863,
      "step": 9915
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.464642724916698e-06,
      "loss": 1.2686,
      "step": 9916
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.463161791928916e-06,
      "loss": 1.3303,
      "step": 9917
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.461680858941133e-06,
      "loss": 1.0797,
      "step": 9918
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.460199925953351e-06,
      "loss": 1.4284,
      "step": 9919
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.4587189929655695e-06,
      "loss": 1.3156,
      "step": 9920
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.457238059977786e-06,
      "loss": 1.0432,
      "step": 9921
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.4557571269900045e-06,
      "loss": 1.1933,
      "step": 9922
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.454276194002221e-06,
      "loss": 1.1779,
      "step": 9923
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.4527952610144396e-06,
      "loss": 1.1595,
      "step": 9924
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.451314328026657e-06,
      "loss": 1.0964,
      "step": 9925
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.449833395038875e-06,
      "loss": 1.2813,
      "step": 9926
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.448352462051093e-06,
      "loss": 1.1054,
      "step": 9927
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.4468715290633105e-06,
      "loss": 1.1289,
      "step": 9928
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.445390596075528e-06,
      "loss": 1.4795,
      "step": 9929
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.4439096630877455e-06,
      "loss": 1.2255,
      "step": 9930
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.442428730099963e-06,
      "loss": 1.2196,
      "step": 9931
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.440947797112181e-06,
      "loss": 1.3799,
      "step": 9932
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.439466864124399e-06,
      "loss": 1.2872,
      "step": 9933
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.4379859311366164e-06,
      "loss": 1.3519,
      "step": 9934
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.436504998148834e-06,
      "loss": 1.0088,
      "step": 9935
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.435024065161052e-06,
      "loss": 1.3122,
      "step": 9936
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.43354313217327e-06,
      "loss": 1.1936,
      "step": 9937
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.432062199185487e-06,
      "loss": 1.1863,
      "step": 9938
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.430581266197706e-06,
      "loss": 1.2261,
      "step": 9939
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.429100333209922e-06,
      "loss": 1.3687,
      "step": 9940
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.427619400222141e-06,
      "loss": 1.2173,
      "step": 9941
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.426138467234357e-06,
      "loss": 1.2571,
      "step": 9942
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.424657534246576e-06,
      "loss": 1.1459,
      "step": 9943
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.423176601258794e-06,
      "loss": 1.3854,
      "step": 9944
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.421695668271011e-06,
      "loss": 1.0945,
      "step": 9945
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.420214735283229e-06,
      "loss": 1.0899,
      "step": 9946
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.418733802295446e-06,
      "loss": 1.2028,
      "step": 9947
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.417252869307664e-06,
      "loss": 1.2804,
      "step": 9948
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.4157719363198826e-06,
      "loss": 1.1872,
      "step": 9949
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.414291003332099e-06,
      "loss": 0.9936,
      "step": 9950
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.412810070344318e-06,
      "loss": 0.9978,
      "step": 9951
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.411329137356534e-06,
      "loss": 1.1087,
      "step": 9952
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.409848204368753e-06,
      "loss": 1.3078,
      "step": 9953
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.408367271380971e-06,
      "loss": 1.1157,
      "step": 9954
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.406886338393188e-06,
      "loss": 1.2684,
      "step": 9955
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.405405405405406e-06,
      "loss": 0.9751,
      "step": 9956
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.4039244724176235e-06,
      "loss": 1.1333,
      "step": 9957
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.402443539429841e-06,
      "loss": 1.0788,
      "step": 9958
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.400962606442059e-06,
      "loss": 1.4528,
      "step": 9959
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.399481673454277e-06,
      "loss": 1.0449,
      "step": 9960
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.3980007404664945e-06,
      "loss": 1.2689,
      "step": 9961
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.396519807478712e-06,
      "loss": 1.0841,
      "step": 9962
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.3950388744909295e-06,
      "loss": 1.1588,
      "step": 9963
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.393557941503147e-06,
      "loss": 1.0879,
      "step": 9964
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.392077008515365e-06,
      "loss": 1.2389,
      "step": 9965
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.390596075527583e-06,
      "loss": 1.202,
      "step": 9966
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.3891151425398e-06,
      "loss": 1.1985,
      "step": 9967
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.387634209552019e-06,
      "loss": 1.1953,
      "step": 9968
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.3861532765642354e-06,
      "loss": 1.099,
      "step": 9969
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.384672343576454e-06,
      "loss": 1.126,
      "step": 9970
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.383191410588672e-06,
      "loss": 1.2271,
      "step": 9971
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.381710477600889e-06,
      "loss": 0.952,
      "step": 9972
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.380229544613107e-06,
      "loss": 1.0767,
      "step": 9973
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.378748611625324e-06,
      "loss": 1.2087,
      "step": 9974
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.377267678637542e-06,
      "loss": 1.1336,
      "step": 9975
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.375786745649761e-06,
      "loss": 1.0324,
      "step": 9976
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.374305812661977e-06,
      "loss": 1.1876,
      "step": 9977
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.372824879674196e-06,
      "loss": 1.208,
      "step": 9978
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.371343946686412e-06,
      "loss": 1.4017,
      "step": 9979
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.369863013698631e-06,
      "loss": 1.0487,
      "step": 9980
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.368382080710848e-06,
      "loss": 1.2973,
      "step": 9981
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.366901147723066e-06,
      "loss": 1.0425,
      "step": 9982
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.365420214735284e-06,
      "loss": 1.1236,
      "step": 9983
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.363939281747501e-06,
      "loss": 1.3484,
      "step": 9984
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.362458348759719e-06,
      "loss": 1.2927,
      "step": 9985
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.360977415771937e-06,
      "loss": 1.2898,
      "step": 9986
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.359496482784154e-06,
      "loss": 1.1599,
      "step": 9987
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.3580155497963725e-06,
      "loss": 1.0231,
      "step": 9988
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.35653461680859e-06,
      "loss": 1.0404,
      "step": 9989
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.3550536838208075e-06,
      "loss": 1.4451,
      "step": 9990
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.353572750833025e-06,
      "loss": 1.2421,
      "step": 9991
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.352091817845243e-06,
      "loss": 1.0864,
      "step": 9992
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.350610884857461e-06,
      "loss": 1.1697,
      "step": 9993
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.3491299518696784e-06,
      "loss": 1.1608,
      "step": 9994
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.347649018881896e-06,
      "loss": 1.173,
      "step": 9995
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.3461680858941135e-06,
      "loss": 1.2638,
      "step": 9996
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.344687152906332e-06,
      "loss": 1.3652,
      "step": 9997
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.3432062199185485e-06,
      "loss": 1.0533,
      "step": 9998
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.341725286930767e-06,
      "loss": 1.1796,
      "step": 9999
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.340244353942985e-06,
      "loss": 1.2273,
      "step": 10000
    },
    {
      "epoch": 2.21,
      "eval_loss": NaN,
      "eval_runtime": 4.378,
      "eval_samples_per_second": 352.671,
      "eval_steps_per_second": 44.084,
      "step": 10000
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.338763420955202e-06,
      "loss": 1.2223,
      "step": 10001
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.33728248796742e-06,
      "loss": 1.1111,
      "step": 10002
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.335801554979637e-06,
      "loss": 1.1679,
      "step": 10003
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.334320621991855e-06,
      "loss": 1.1991,
      "step": 10004
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.332839689004074e-06,
      "loss": 1.1431,
      "step": 10005
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.33135875601629e-06,
      "loss": 1.1539,
      "step": 10006
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.329877823028509e-06,
      "loss": 1.2026,
      "step": 10007
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.328396890040725e-06,
      "loss": 1.2049,
      "step": 10008
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.326915957052944e-06,
      "loss": 0.8881,
      "step": 10009
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.325435024065162e-06,
      "loss": 1.3505,
      "step": 10010
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.323954091077379e-06,
      "loss": 1.3585,
      "step": 10011
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.322473158089597e-06,
      "loss": 1.0291,
      "step": 10012
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.320992225101815e-06,
      "loss": 1.4224,
      "step": 10013
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.319511292114032e-06,
      "loss": 1.2496,
      "step": 10014
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.31803035912625e-06,
      "loss": 1.2561,
      "step": 10015
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.316549426138467e-06,
      "loss": 1.4434,
      "step": 10016
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.3150684931506856e-06,
      "loss": 1.2314,
      "step": 10017
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.313587560162903e-06,
      "loss": 1.0795,
      "step": 10018
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.312106627175121e-06,
      "loss": 1.3815,
      "step": 10019
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.310625694187338e-06,
      "loss": 1.3623,
      "step": 10020
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.3091447611995565e-06,
      "loss": 1.1429,
      "step": 10021
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.307663828211774e-06,
      "loss": 1.0333,
      "step": 10022
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.3061828952239915e-06,
      "loss": 1.2363,
      "step": 10023
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.30470196223621e-06,
      "loss": 1.0897,
      "step": 10024
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.3032210292484266e-06,
      "loss": 1.2412,
      "step": 10025
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.301740096260645e-06,
      "loss": 1.1953,
      "step": 10026
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.3002591632728624e-06,
      "loss": 1.245,
      "step": 10027
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.29877823028508e-06,
      "loss": 1.2058,
      "step": 10028
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.297297297297298e-06,
      "loss": 1.167,
      "step": 10029
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.295816364309515e-06,
      "loss": 1.306,
      "step": 10030
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.294335431321733e-06,
      "loss": 1.1735,
      "step": 10031
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.292854498333952e-06,
      "loss": 1.3066,
      "step": 10032
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.291373565346168e-06,
      "loss": 1.2342,
      "step": 10033
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.289892632358387e-06,
      "loss": 1.3931,
      "step": 10034
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.288411699370603e-06,
      "loss": 1.2772,
      "step": 10035
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.286930766382822e-06,
      "loss": 1.3338,
      "step": 10036
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.2854498333950385e-06,
      "loss": 1.2122,
      "step": 10037
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.283968900407257e-06,
      "loss": 1.1175,
      "step": 10038
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.282487967419475e-06,
      "loss": 1.2771,
      "step": 10039
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.281007034431692e-06,
      "loss": 1.215,
      "step": 10040
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.27952610144391e-06,
      "loss": 1.1093,
      "step": 10041
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.278045168456128e-06,
      "loss": 1.1296,
      "step": 10042
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.276564235468345e-06,
      "loss": 1.2916,
      "step": 10043
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.275083302480564e-06,
      "loss": 1.4479,
      "step": 10044
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.27360236949278e-06,
      "loss": 1.0627,
      "step": 10045
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.272121436504999e-06,
      "loss": 1.1592,
      "step": 10046
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.270640503517216e-06,
      "loss": 1.3363,
      "step": 10047
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.269159570529434e-06,
      "loss": 1.0895,
      "step": 10048
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.267678637541652e-06,
      "loss": 1.1524,
      "step": 10049
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.2661977045538696e-06,
      "loss": 1.2299,
      "step": 10050
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.264716771566087e-06,
      "loss": 1.088,
      "step": 10051
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.263235838578305e-06,
      "loss": 1.2631,
      "step": 10052
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.261754905590523e-06,
      "loss": 1.3062,
      "step": 10053
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.26027397260274e-06,
      "loss": 1.1791,
      "step": 10054
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.258793039614958e-06,
      "loss": 1.3592,
      "step": 10055
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.257312106627176e-06,
      "loss": 1.2168,
      "step": 10056
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.255831173639393e-06,
      "loss": 1.2931,
      "step": 10057
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.254350240651611e-06,
      "loss": 1.2338,
      "step": 10058
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.252869307663828e-06,
      "loss": 1.1798,
      "step": 10059
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.251388374676046e-06,
      "loss": 1.4517,
      "step": 10060
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.249907441688265e-06,
      "loss": 1.1409,
      "step": 10061
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.2484265087004815e-06,
      "loss": 1.12,
      "step": 10062
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.2469455757127e-06,
      "loss": 1.1265,
      "step": 10063
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.2454646427249165e-06,
      "loss": 1.2039,
      "step": 10064
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.243983709737135e-06,
      "loss": 1.1346,
      "step": 10065
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.242502776749353e-06,
      "loss": 1.0931,
      "step": 10066
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.24102184376157e-06,
      "loss": 1.0266,
      "step": 10067
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.239540910773788e-06,
      "loss": 1.2044,
      "step": 10068
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.238059977786005e-06,
      "loss": 1.209,
      "step": 10069
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.236579044798223e-06,
      "loss": 1.3529,
      "step": 10070
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.235098111810441e-06,
      "loss": 1.5347,
      "step": 10071
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.233617178822658e-06,
      "loss": 1.004,
      "step": 10072
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.232136245834877e-06,
      "loss": 1.1702,
      "step": 10073
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.230655312847094e-06,
      "loss": 1.3276,
      "step": 10074
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.229174379859312e-06,
      "loss": 1.2446,
      "step": 10075
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.227693446871529e-06,
      "loss": 1.3678,
      "step": 10076
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.226212513883747e-06,
      "loss": 1.3478,
      "step": 10077
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.224731580895965e-06,
      "loss": 1.3309,
      "step": 10078
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.223250647908183e-06,
      "loss": 1.4942,
      "step": 10079
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.2217697149204e-06,
      "loss": 1.2057,
      "step": 10080
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.220288781932618e-06,
      "loss": 1.1975,
      "step": 10081
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.218807848944836e-06,
      "loss": 1.0768,
      "step": 10082
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.2173269159570535e-06,
      "loss": 1.3246,
      "step": 10083
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.215845982969271e-06,
      "loss": 1.104,
      "step": 10084
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.2143650499814894e-06,
      "loss": 1.1718,
      "step": 10085
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.212884116993706e-06,
      "loss": 1.1296,
      "step": 10086
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.2114031840059245e-06,
      "loss": 1.1381,
      "step": 10087
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.209922251018141e-06,
      "loss": 1.4195,
      "step": 10088
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.2084413180303595e-06,
      "loss": 1.4901,
      "step": 10089
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.206960385042578e-06,
      "loss": 1.1185,
      "step": 10090
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.2054794520547945e-06,
      "loss": 1.3361,
      "step": 10091
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.203998519067013e-06,
      "loss": 1.2709,
      "step": 10092
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.2025175860792296e-06,
      "loss": 1.2503,
      "step": 10093
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.201036653091448e-06,
      "loss": 1.2235,
      "step": 10094
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.199555720103666e-06,
      "loss": 1.4074,
      "step": 10095
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.198074787115883e-06,
      "loss": 1.1083,
      "step": 10096
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.196593854128101e-06,
      "loss": 1.3574,
      "step": 10097
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.195112921140318e-06,
      "loss": 1.2209,
      "step": 10098
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.193631988152536e-06,
      "loss": 1.171,
      "step": 10099
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.192151055164755e-06,
      "loss": 1.1069,
      "step": 10100
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.190670122176971e-06,
      "loss": 1.131,
      "step": 10101
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.18918918918919e-06,
      "loss": 1.108,
      "step": 10102
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.187708256201407e-06,
      "loss": 1.1692,
      "step": 10103
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.186227323213625e-06,
      "loss": 1.4669,
      "step": 10104
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.184746390225843e-06,
      "loss": 1.2859,
      "step": 10105
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.183265457238061e-06,
      "loss": 1.2508,
      "step": 10106
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.181784524250278e-06,
      "loss": 1.0712,
      "step": 10107
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.180303591262496e-06,
      "loss": 1.111,
      "step": 10108
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.178822658274713e-06,
      "loss": 1.0946,
      "step": 10109
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.177341725286931e-06,
      "loss": 0.9707,
      "step": 10110
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.175860792299149e-06,
      "loss": 1.4026,
      "step": 10111
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.174379859311367e-06,
      "loss": 1.2862,
      "step": 10112
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.172898926323584e-06,
      "loss": 1.1819,
      "step": 10113
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.1714179933358025e-06,
      "loss": 1.1904,
      "step": 10114
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.169937060348019e-06,
      "loss": 1.2017,
      "step": 10115
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.1684561273602375e-06,
      "loss": 1.3017,
      "step": 10116
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.166975194372456e-06,
      "loss": 1.2113,
      "step": 10117
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.1654942613846726e-06,
      "loss": 1.3276,
      "step": 10118
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.164013328396891e-06,
      "loss": 1.1664,
      "step": 10119
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.162532395409108e-06,
      "loss": 1.1701,
      "step": 10120
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.161051462421326e-06,
      "loss": 1.1326,
      "step": 10121
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.159570529433544e-06,
      "loss": 0.9391,
      "step": 10122
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.158089596445761e-06,
      "loss": 1.1663,
      "step": 10123
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.156608663457979e-06,
      "loss": 1.2586,
      "step": 10124
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.155127730470196e-06,
      "loss": 1.315,
      "step": 10125
    },
    {
      "epoch": 2.23,
      "eval_loss": NaN,
      "eval_runtime": 4.3769,
      "eval_samples_per_second": 352.761,
      "eval_steps_per_second": 44.095,
      "step": 10125
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.153646797482414e-06,
      "loss": 1.4563,
      "step": 10126
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.152165864494632e-06,
      "loss": 1.3523,
      "step": 10127
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.1506849315068494e-06,
      "loss": 1.2782,
      "step": 10128
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.149203998519068e-06,
      "loss": 1.2536,
      "step": 10129
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.1477230655312845e-06,
      "loss": 1.2002,
      "step": 10130
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.146242132543503e-06,
      "loss": 1.5503,
      "step": 10131
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.14476119955572e-06,
      "loss": 1.1874,
      "step": 10132
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.143280266567938e-06,
      "loss": 1.2735,
      "step": 10133
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.141799333580156e-06,
      "loss": 1.2228,
      "step": 10134
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.140318400592374e-06,
      "loss": 1.3749,
      "step": 10135
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.138837467604591e-06,
      "loss": 1.131,
      "step": 10136
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.137356534616809e-06,
      "loss": 1.0447,
      "step": 10137
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.135875601629027e-06,
      "loss": 1.1491,
      "step": 10138
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.134394668641245e-06,
      "loss": 1.3221,
      "step": 10139
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.132913735653462e-06,
      "loss": 1.2577,
      "step": 10140
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.13143280266568e-06,
      "loss": 1.3872,
      "step": 10141
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.129951869677897e-06,
      "loss": 0.9944,
      "step": 10142
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.1284709366901156e-06,
      "loss": 1.1638,
      "step": 10143
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.126990003702332e-06,
      "loss": 1.153,
      "step": 10144
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.125509070714551e-06,
      "loss": 1.1425,
      "step": 10145
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.124028137726769e-06,
      "loss": 1.2744,
      "step": 10146
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.122547204738986e-06,
      "loss": 1.0827,
      "step": 10147
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.121066271751204e-06,
      "loss": 1.2029,
      "step": 10148
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.119585338763421e-06,
      "loss": 1.1881,
      "step": 10149
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.118104405775639e-06,
      "loss": 1.0452,
      "step": 10150
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.116623472787857e-06,
      "loss": 1.0467,
      "step": 10151
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.115142539800074e-06,
      "loss": 0.9199,
      "step": 10152
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.1136616068122924e-06,
      "loss": 0.9897,
      "step": 10153
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.112180673824509e-06,
      "loss": 1.2467,
      "step": 10154
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.1106997408367275e-06,
      "loss": 1.4407,
      "step": 10155
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.109218807848946e-06,
      "loss": 1.2039,
      "step": 10156
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.1077378748611625e-06,
      "loss": 1.1301,
      "step": 10157
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.106256941873381e-06,
      "loss": 1.1776,
      "step": 10158
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.104776008885598e-06,
      "loss": 1.3744,
      "step": 10159
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.103295075897816e-06,
      "loss": 1.2726,
      "step": 10160
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.101814142910033e-06,
      "loss": 1.1687,
      "step": 10161
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.100333209922251e-06,
      "loss": 1.156,
      "step": 10162
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.098852276934469e-06,
      "loss": 1.2861,
      "step": 10163
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.097371343946687e-06,
      "loss": 1.1021,
      "step": 10164
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.095890410958904e-06,
      "loss": 1.2175,
      "step": 10165
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.094409477971122e-06,
      "loss": 1.2941,
      "step": 10166
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.09292854498334e-06,
      "loss": 1.0222,
      "step": 10167
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.091447611995558e-06,
      "loss": 1.1583,
      "step": 10168
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.089966679007775e-06,
      "loss": 1.2506,
      "step": 10169
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.088485746019994e-06,
      "loss": 1.4511,
      "step": 10170
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.08700481303221e-06,
      "loss": 1.3088,
      "step": 10171
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.085523880044429e-06,
      "loss": 1.1509,
      "step": 10172
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.084042947056646e-06,
      "loss": 1.1468,
      "step": 10173
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.082562014068864e-06,
      "loss": 1.3699,
      "step": 10174
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.081081081081082e-06,
      "loss": 1.1485,
      "step": 10175
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.079600148093299e-06,
      "loss": 1.1166,
      "step": 10176
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.078119215105517e-06,
      "loss": 1.3078,
      "step": 10177
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.0766382821177354e-06,
      "loss": 1.2102,
      "step": 10178
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.075157349129952e-06,
      "loss": 1.1436,
      "step": 10179
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.0736764161421705e-06,
      "loss": 1.4182,
      "step": 10180
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.072195483154387e-06,
      "loss": 1.1932,
      "step": 10181
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.0707145501666055e-06,
      "loss": 1.4379,
      "step": 10182
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.069233617178822e-06,
      "loss": 1.1996,
      "step": 10183
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.0677526841910405e-06,
      "loss": 1.1704,
      "step": 10184
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.066271751203259e-06,
      "loss": 0.9738,
      "step": 10185
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.0647908182154756e-06,
      "loss": 1.1562,
      "step": 10186
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.063309885227694e-06,
      "loss": 1.0662,
      "step": 10187
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.0618289522399115e-06,
      "loss": 1.095,
      "step": 10188
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.060348019252129e-06,
      "loss": 1.2712,
      "step": 10189
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.058867086264347e-06,
      "loss": 1.0884,
      "step": 10190
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.057386153276565e-06,
      "loss": 1.3328,
      "step": 10191
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.055905220288782e-06,
      "loss": 1.184,
      "step": 10192
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.054424287301e-06,
      "loss": 1.3097,
      "step": 10193
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.052943354313217e-06,
      "loss": 1.1174,
      "step": 10194
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.051462421325436e-06,
      "loss": 1.2611,
      "step": 10195
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.049981488337653e-06,
      "loss": 1.17,
      "step": 10196
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.048500555349871e-06,
      "loss": 1.5275,
      "step": 10197
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.047019622362088e-06,
      "loss": 1.2538,
      "step": 10198
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.045538689374307e-06,
      "loss": 1.0773,
      "step": 10199
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.044057756386523e-06,
      "loss": 1.3346,
      "step": 10200
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.042576823398742e-06,
      "loss": 1.0993,
      "step": 10201
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.04109589041096e-06,
      "loss": 1.4486,
      "step": 10202
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.039614957423177e-06,
      "loss": 1.2787,
      "step": 10203
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.038134024435395e-06,
      "loss": 1.3044,
      "step": 10204
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.036653091447612e-06,
      "loss": 1.3712,
      "step": 10205
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.03517215845983e-06,
      "loss": 1.3176,
      "step": 10206
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.0336912254720485e-06,
      "loss": 1.2497,
      "step": 10207
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.032210292484265e-06,
      "loss": 1.2395,
      "step": 10208
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.0307293594964835e-06,
      "loss": 1.2635,
      "step": 10209
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.0292484265087e-06,
      "loss": 1.3345,
      "step": 10210
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.027767493520919e-06,
      "loss": 1.4251,
      "step": 10211
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.026286560533137e-06,
      "loss": 0.9865,
      "step": 10212
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.024805627545354e-06,
      "loss": 1.236,
      "step": 10213
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.023324694557572e-06,
      "loss": 1.3086,
      "step": 10214
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.021843761569789e-06,
      "loss": 1.2105,
      "step": 10215
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.020362828582007e-06,
      "loss": 1.2403,
      "step": 10216
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.0188818955942245e-06,
      "loss": 1.164,
      "step": 10217
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.017400962606442e-06,
      "loss": 1.1688,
      "step": 10218
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.01592002961866e-06,
      "loss": 1.0136,
      "step": 10219
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.014439096630878e-06,
      "loss": 1.1379,
      "step": 10220
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.0129581636430954e-06,
      "loss": 1.2,
      "step": 10221
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.011477230655313e-06,
      "loss": 1.1016,
      "step": 10222
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.009996297667531e-06,
      "loss": 1.2116,
      "step": 10223
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.008515364679749e-06,
      "loss": 1.2064,
      "step": 10224
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.007034431691966e-06,
      "loss": 1.2495,
      "step": 10225
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.005553498704184e-06,
      "loss": 1.305,
      "step": 10226
    },
    {
      "epoch": 2.26,
      "learning_rate": 5.004072565716401e-06,
      "loss": 1.3367,
      "step": 10227
    },
    {
      "epoch": 2.26,
      "learning_rate": 5.00259163272862e-06,
      "loss": 1.2919,
      "step": 10228
    },
    {
      "epoch": 2.26,
      "learning_rate": 5.001110699740837e-06,
      "loss": 1.3727,
      "step": 10229
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.999629766753055e-06,
      "loss": 1.3026,
      "step": 10230
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.998148833765272e-06,
      "loss": 1.136,
      "step": 10231
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.996667900777491e-06,
      "loss": 1.4462,
      "step": 10232
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.995186967789708e-06,
      "loss": 1.1292,
      "step": 10233
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.993706034801926e-06,
      "loss": 1.2974,
      "step": 10234
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.992225101814143e-06,
      "loss": 1.1532,
      "step": 10235
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.990744168826361e-06,
      "loss": 1.1018,
      "step": 10236
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.989263235838579e-06,
      "loss": 1.1167,
      "step": 10237
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.987782302850797e-06,
      "loss": 1.3177,
      "step": 10238
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.986301369863014e-06,
      "loss": 1.1718,
      "step": 10239
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.984820436875232e-06,
      "loss": 1.1739,
      "step": 10240
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.983339503887449e-06,
      "loss": 1.1238,
      "step": 10241
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.981858570899667e-06,
      "loss": 1.0906,
      "step": 10242
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.980377637911885e-06,
      "loss": 1.2587,
      "step": 10243
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.9788967049241026e-06,
      "loss": 1.216,
      "step": 10244
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.97741577193632e-06,
      "loss": 1.2929,
      "step": 10245
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.975934838948538e-06,
      "loss": 1.1149,
      "step": 10246
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.974453905960755e-06,
      "loss": 1.238,
      "step": 10247
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.9729729729729735e-06,
      "loss": 1.1839,
      "step": 10248
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.971492039985191e-06,
      "loss": 1.1336,
      "step": 10249
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.9700111069974085e-06,
      "loss": 0.9986,
      "step": 10250
    },
    {
      "epoch": 2.26,
      "eval_loss": NaN,
      "eval_runtime": 4.369,
      "eval_samples_per_second": 353.402,
      "eval_steps_per_second": 44.175,
      "step": 10250
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.968530174009626e-06,
      "loss": 1.0691,
      "step": 10251
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.967049241021844e-06,
      "loss": 1.3141,
      "step": 10252
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.965568308034062e-06,
      "loss": 1.0944,
      "step": 10253
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.9640873750462794e-06,
      "loss": 1.3064,
      "step": 10254
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.962606442058498e-06,
      "loss": 1.266,
      "step": 10255
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.961125509070715e-06,
      "loss": 1.1071,
      "step": 10256
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.959644576082933e-06,
      "loss": 1.4703,
      "step": 10257
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.95816364309515e-06,
      "loss": 1.2519,
      "step": 10258
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.956682710107368e-06,
      "loss": 1.2789,
      "step": 10259
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.955201777119586e-06,
      "loss": 1.3665,
      "step": 10260
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.953720844131804e-06,
      "loss": 1.1778,
      "step": 10261
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.952239911144021e-06,
      "loss": 1.3314,
      "step": 10262
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.950758978156239e-06,
      "loss": 1.1873,
      "step": 10263
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.949278045168456e-06,
      "loss": 1.129,
      "step": 10264
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.947797112180675e-06,
      "loss": 1.1927,
      "step": 10265
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.946316179192892e-06,
      "loss": 1.3398,
      "step": 10266
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.94483524620511e-06,
      "loss": 1.1958,
      "step": 10267
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.943354313217327e-06,
      "loss": 1.3227,
      "step": 10268
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.941873380229545e-06,
      "loss": 1.3279,
      "step": 10269
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.940392447241762e-06,
      "loss": 1.3134,
      "step": 10270
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.938911514253981e-06,
      "loss": 1.1179,
      "step": 10271
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.937430581266198e-06,
      "loss": 1.3129,
      "step": 10272
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.935949648278416e-06,
      "loss": 1.1932,
      "step": 10273
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.934468715290633e-06,
      "loss": 1.2502,
      "step": 10274
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.932987782302851e-06,
      "loss": 1.1035,
      "step": 10275
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.931506849315069e-06,
      "loss": 1.2933,
      "step": 10276
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.9300259163272866e-06,
      "loss": 1.0889,
      "step": 10277
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.928544983339504e-06,
      "loss": 1.1458,
      "step": 10278
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.927064050351722e-06,
      "loss": 1.2591,
      "step": 10279
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.92558311736394e-06,
      "loss": 1.088,
      "step": 10280
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.9241021843761575e-06,
      "loss": 1.3574,
      "step": 10281
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.922621251388375e-06,
      "loss": 0.9924,
      "step": 10282
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.9211403184005925e-06,
      "loss": 1.0621,
      "step": 10283
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.919659385412811e-06,
      "loss": 1.3063,
      "step": 10284
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.918178452425028e-06,
      "loss": 1.3214,
      "step": 10285
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.916697519437246e-06,
      "loss": 1.2188,
      "step": 10286
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.915216586449463e-06,
      "loss": 1.2709,
      "step": 10287
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.913735653461682e-06,
      "loss": 1.0049,
      "step": 10288
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.912254720473899e-06,
      "loss": 1.1606,
      "step": 10289
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.910773787486117e-06,
      "loss": 1.056,
      "step": 10290
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.909292854498334e-06,
      "loss": 1.4925,
      "step": 10291
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.907811921510552e-06,
      "loss": 1.3082,
      "step": 10292
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.90633098852277e-06,
      "loss": 1.6334,
      "step": 10293
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.904850055534988e-06,
      "loss": 1.152,
      "step": 10294
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.903369122547205e-06,
      "loss": 1.2054,
      "step": 10295
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.901888189559423e-06,
      "loss": 1.0974,
      "step": 10296
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.90040725657164e-06,
      "loss": 1.1065,
      "step": 10297
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.898926323583858e-06,
      "loss": 1.4613,
      "step": 10298
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.897445390596076e-06,
      "loss": 1.2186,
      "step": 10299
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.895964457608294e-06,
      "loss": 1.1605,
      "step": 10300
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.894483524620511e-06,
      "loss": 1.0926,
      "step": 10301
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.893002591632729e-06,
      "loss": 1.3318,
      "step": 10302
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.891521658644946e-06,
      "loss": 1.1298,
      "step": 10303
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.890040725657164e-06,
      "loss": 1.0864,
      "step": 10304
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.888559792669382e-06,
      "loss": 1.2843,
      "step": 10305
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.8870788596816e-06,
      "loss": 1.3073,
      "step": 10306
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.885597926693817e-06,
      "loss": 1.3157,
      "step": 10307
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.8841169937060355e-06,
      "loss": 1.2319,
      "step": 10308
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.882636060718253e-06,
      "loss": 1.3451,
      "step": 10309
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.8811551277304705e-06,
      "loss": 1.286,
      "step": 10310
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.879674194742688e-06,
      "loss": 1.1989,
      "step": 10311
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.878193261754906e-06,
      "loss": 1.1732,
      "step": 10312
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.876712328767124e-06,
      "loss": 1.2561,
      "step": 10313
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.8752313957793415e-06,
      "loss": 1.2347,
      "step": 10314
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.873750462791559e-06,
      "loss": 1.1715,
      "step": 10315
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.872269529803777e-06,
      "loss": 1.1663,
      "step": 10316
    },
    {
      "epoch": 2.27,
      "learning_rate": 4.870788596815995e-06,
      "loss": 1.3229,
      "step": 10317
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.869307663828212e-06,
      "loss": 1.1963,
      "step": 10318
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.86782673084043e-06,
      "loss": 1.118,
      "step": 10319
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.866345797852647e-06,
      "loss": 0.9427,
      "step": 10320
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.864864864864866e-06,
      "loss": 1.1524,
      "step": 10321
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.863383931877083e-06,
      "loss": 1.3346,
      "step": 10322
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.861902998889301e-06,
      "loss": 1.316,
      "step": 10323
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.860422065901518e-06,
      "loss": 1.1823,
      "step": 10324
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.858941132913736e-06,
      "loss": 1.5406,
      "step": 10325
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.857460199925953e-06,
      "loss": 1.2388,
      "step": 10326
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.855979266938172e-06,
      "loss": 1.1297,
      "step": 10327
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.854498333950389e-06,
      "loss": 1.0661,
      "step": 10328
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.853017400962607e-06,
      "loss": 1.2313,
      "step": 10329
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.851536467974824e-06,
      "loss": 1.0249,
      "step": 10330
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.850055534987042e-06,
      "loss": 1.2123,
      "step": 10331
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.848574601999259e-06,
      "loss": 1.2011,
      "step": 10332
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.847093669011478e-06,
      "loss": 1.1931,
      "step": 10333
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.845612736023695e-06,
      "loss": 1.2913,
      "step": 10334
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.844131803035913e-06,
      "loss": 1.4796,
      "step": 10335
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.84265087004813e-06,
      "loss": 1.396,
      "step": 10336
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.841169937060349e-06,
      "loss": 1.3736,
      "step": 10337
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.839689004072566e-06,
      "loss": 0.9986,
      "step": 10338
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.838208071084784e-06,
      "loss": 1.2185,
      "step": 10339
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.836727138097002e-06,
      "loss": 1.327,
      "step": 10340
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.8352462051092195e-06,
      "loss": 1.3687,
      "step": 10341
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.833765272121437e-06,
      "loss": 1.091,
      "step": 10342
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.8322843391336545e-06,
      "loss": 1.1895,
      "step": 10343
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.830803406145873e-06,
      "loss": 1.1601,
      "step": 10344
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.82932247315809e-06,
      "loss": 1.3052,
      "step": 10345
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.827841540170308e-06,
      "loss": 1.192,
      "step": 10346
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.8263606071825254e-06,
      "loss": 1.1211,
      "step": 10347
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.824879674194743e-06,
      "loss": 1.0496,
      "step": 10348
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.823398741206961e-06,
      "loss": 1.0911,
      "step": 10349
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.821917808219179e-06,
      "loss": 1.2715,
      "step": 10350
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.820436875231396e-06,
      "loss": 1.2818,
      "step": 10351
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.818955942243614e-06,
      "loss": 1.3214,
      "step": 10352
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.817475009255831e-06,
      "loss": 1.2221,
      "step": 10353
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.815994076268049e-06,
      "loss": 1.1931,
      "step": 10354
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.814513143280267e-06,
      "loss": 1.3167,
      "step": 10355
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.813032210292485e-06,
      "loss": 1.1436,
      "step": 10356
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.811551277304702e-06,
      "loss": 1.1882,
      "step": 10357
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.81007034431692e-06,
      "loss": 1.3221,
      "step": 10358
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.808589411329137e-06,
      "loss": 1.2515,
      "step": 10359
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.807108478341355e-06,
      "loss": 1.1814,
      "step": 10360
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.805627545353573e-06,
      "loss": 1.1956,
      "step": 10361
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.804146612365791e-06,
      "loss": 1.2314,
      "step": 10362
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.802665679378008e-06,
      "loss": 1.3587,
      "step": 10363
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.801184746390226e-06,
      "loss": 1.2184,
      "step": 10364
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.799703813402444e-06,
      "loss": 1.4051,
      "step": 10365
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.798222880414662e-06,
      "loss": 1.2185,
      "step": 10366
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.796741947426879e-06,
      "loss": 1.003,
      "step": 10367
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.795261014439097e-06,
      "loss": 1.1661,
      "step": 10368
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.793780081451315e-06,
      "loss": 1.2354,
      "step": 10369
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.7922991484635326e-06,
      "loss": 1.2224,
      "step": 10370
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.79081821547575e-06,
      "loss": 1.0571,
      "step": 10371
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.7893372824879684e-06,
      "loss": 1.3346,
      "step": 10372
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.787856349500186e-06,
      "loss": 1.1863,
      "step": 10373
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.7863754165124035e-06,
      "loss": 1.4066,
      "step": 10374
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.784894483524621e-06,
      "loss": 1.1584,
      "step": 10375
    },
    {
      "epoch": 2.29,
      "eval_loss": NaN,
      "eval_runtime": 4.3742,
      "eval_samples_per_second": 352.98,
      "eval_steps_per_second": 44.123,
      "step": 10375
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.7834135505368385e-06,
      "loss": 1.312,
      "step": 10376
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.781932617549056e-06,
      "loss": 0.9176,
      "step": 10377
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.780451684561274e-06,
      "loss": 1.1056,
      "step": 10378
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.778970751573492e-06,
      "loss": 1.1763,
      "step": 10379
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.7774898185857094e-06,
      "loss": 1.2586,
      "step": 10380
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.776008885597927e-06,
      "loss": 1.1686,
      "step": 10381
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.7745279526101445e-06,
      "loss": 1.0199,
      "step": 10382
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.773047019622363e-06,
      "loss": 1.3055,
      "step": 10383
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.77156608663458e-06,
      "loss": 1.4323,
      "step": 10384
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.770085153646798e-06,
      "loss": 1.3778,
      "step": 10385
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.768604220659015e-06,
      "loss": 1.2343,
      "step": 10386
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.767123287671233e-06,
      "loss": 1.1596,
      "step": 10387
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.76564235468345e-06,
      "loss": 1.4,
      "step": 10388
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.764161421695669e-06,
      "loss": 1.3319,
      "step": 10389
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.762680488707886e-06,
      "loss": 1.2356,
      "step": 10390
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.761199555720104e-06,
      "loss": 1.3079,
      "step": 10391
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.759718622732321e-06,
      "loss": 1.0589,
      "step": 10392
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.758237689744539e-06,
      "loss": 1.2113,
      "step": 10393
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.756756756756757e-06,
      "loss": 1.3166,
      "step": 10394
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.755275823768975e-06,
      "loss": 1.4794,
      "step": 10395
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.753794890781192e-06,
      "loss": 1.0492,
      "step": 10396
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.752313957793411e-06,
      "loss": 1.1707,
      "step": 10397
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.750833024805628e-06,
      "loss": 1.2362,
      "step": 10398
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.749352091817846e-06,
      "loss": 1.3402,
      "step": 10399
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.747871158830063e-06,
      "loss": 1.3286,
      "step": 10400
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.7463902258422815e-06,
      "loss": 1.1302,
      "step": 10401
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.744909292854499e-06,
      "loss": 1.1943,
      "step": 10402
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.7434283598667166e-06,
      "loss": 1.4591,
      "step": 10403
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.741947426878934e-06,
      "loss": 1.3046,
      "step": 10404
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.740466493891152e-06,
      "loss": 1.4148,
      "step": 10405
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.73898556090337e-06,
      "loss": 1.2105,
      "step": 10406
    },
    {
      "epoch": 2.29,
      "learning_rate": 4.7375046279155875e-06,
      "loss": 1.3824,
      "step": 10407
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.736023694927805e-06,
      "loss": 1.3945,
      "step": 10408
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.7345427619400225e-06,
      "loss": 0.9702,
      "step": 10409
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.73306182895224e-06,
      "loss": 1.1482,
      "step": 10410
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.731580895964458e-06,
      "loss": 1.292,
      "step": 10411
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.730099962976676e-06,
      "loss": 1.239,
      "step": 10412
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.728619029988893e-06,
      "loss": 1.1077,
      "step": 10413
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.727138097001111e-06,
      "loss": 1.291,
      "step": 10414
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.7256571640133284e-06,
      "loss": 1.1668,
      "step": 10415
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.724176231025546e-06,
      "loss": 1.2648,
      "step": 10416
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.722695298037764e-06,
      "loss": 1.0677,
      "step": 10417
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.721214365049982e-06,
      "loss": 1.4763,
      "step": 10418
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.719733432062199e-06,
      "loss": 1.114,
      "step": 10419
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.718252499074417e-06,
      "loss": 1.0336,
      "step": 10420
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.716771566086634e-06,
      "loss": 1.2412,
      "step": 10421
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.715290633098853e-06,
      "loss": 1.2168,
      "step": 10422
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.71380970011107e-06,
      "loss": 1.0684,
      "step": 10423
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.712328767123288e-06,
      "loss": 1.2528,
      "step": 10424
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.710847834135505e-06,
      "loss": 1.3481,
      "step": 10425
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.709366901147724e-06,
      "loss": 1.2159,
      "step": 10426
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.707885968159941e-06,
      "loss": 1.0264,
      "step": 10427
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.706405035172159e-06,
      "loss": 1.2792,
      "step": 10428
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.704924102184377e-06,
      "loss": 1.0326,
      "step": 10429
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.703443169196595e-06,
      "loss": 1.2923,
      "step": 10430
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.701962236208812e-06,
      "loss": 1.1991,
      "step": 10431
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.70048130322103e-06,
      "loss": 1.2936,
      "step": 10432
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.699000370233247e-06,
      "loss": 1.2776,
      "step": 10433
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.6975194372454655e-06,
      "loss": 1.106,
      "step": 10434
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.696038504257683e-06,
      "loss": 1.2896,
      "step": 10435
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.6945575712699005e-06,
      "loss": 1.3846,
      "step": 10436
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.693076638282118e-06,
      "loss": 1.2292,
      "step": 10437
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.6915957052943356e-06,
      "loss": 1.0436,
      "step": 10438
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.690114772306554e-06,
      "loss": 1.1019,
      "step": 10439
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.6886338393187715e-06,
      "loss": 1.0738,
      "step": 10440
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.687152906330989e-06,
      "loss": 1.0778,
      "step": 10441
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.6856719733432065e-06,
      "loss": 1.1718,
      "step": 10442
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.684191040355424e-06,
      "loss": 1.4929,
      "step": 10443
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.6827101073676415e-06,
      "loss": 1.4867,
      "step": 10444
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.68122917437986e-06,
      "loss": 1.1862,
      "step": 10445
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.679748241392077e-06,
      "loss": 1.2442,
      "step": 10446
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.678267308404295e-06,
      "loss": 1.1075,
      "step": 10447
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.6767863754165124e-06,
      "loss": 1.3682,
      "step": 10448
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.67530544242873e-06,
      "loss": 1.1199,
      "step": 10449
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.673824509440948e-06,
      "loss": 1.2604,
      "step": 10450
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.672343576453166e-06,
      "loss": 1.1887,
      "step": 10451
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.670862643465383e-06,
      "loss": 1.0341,
      "step": 10452
    },
    {
      "epoch": 2.3,
      "learning_rate": 4.669381710477601e-06,
      "loss": 1.233,
      "step": 10453
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.667900777489819e-06,
      "loss": 1.1387,
      "step": 10454
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.666419844502037e-06,
      "loss": 1.1326,
      "step": 10455
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.664938911514254e-06,
      "loss": 1.2907,
      "step": 10456
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.663457978526472e-06,
      "loss": 1.2296,
      "step": 10457
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.66197704553869e-06,
      "loss": 1.0425,
      "step": 10458
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.660496112550908e-06,
      "loss": 1.1906,
      "step": 10459
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.659015179563125e-06,
      "loss": 1.2587,
      "step": 10460
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.657534246575343e-06,
      "loss": 1.1967,
      "step": 10461
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.656053313587561e-06,
      "loss": 1.046,
      "step": 10462
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.654572380599779e-06,
      "loss": 1.3737,
      "step": 10463
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.653091447611996e-06,
      "loss": 1.1923,
      "step": 10464
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.651610514624214e-06,
      "loss": 1.1413,
      "step": 10465
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.650129581636431e-06,
      "loss": 1.1054,
      "step": 10466
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.6486486486486495e-06,
      "loss": 1.3188,
      "step": 10467
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.647167715660867e-06,
      "loss": 1.1894,
      "step": 10468
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.6456867826730845e-06,
      "loss": 1.4606,
      "step": 10469
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.644205849685302e-06,
      "loss": 1.099,
      "step": 10470
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.6427249166975196e-06,
      "loss": 1.3108,
      "step": 10471
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.641243983709737e-06,
      "loss": 1.2096,
      "step": 10472
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.6397630507219554e-06,
      "loss": 1.277,
      "step": 10473
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.638282117734173e-06,
      "loss": 1.1897,
      "step": 10474
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.6368011847463905e-06,
      "loss": 1.2575,
      "step": 10475
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.635320251758608e-06,
      "loss": 1.2039,
      "step": 10476
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.6338393187708255e-06,
      "loss": 1.3931,
      "step": 10477
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.632358385783043e-06,
      "loss": 1.1076,
      "step": 10478
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.630877452795261e-06,
      "loss": 1.2587,
      "step": 10479
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.629396519807479e-06,
      "loss": 1.3379,
      "step": 10480
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.627915586819696e-06,
      "loss": 1.1149,
      "step": 10481
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.626434653831915e-06,
      "loss": 1.2614,
      "step": 10482
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.624953720844132e-06,
      "loss": 1.1103,
      "step": 10483
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.62347278785635e-06,
      "loss": 1.3336,
      "step": 10484
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.621991854868567e-06,
      "loss": 1.0638,
      "step": 10485
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.620510921880786e-06,
      "loss": 1.1356,
      "step": 10486
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.619029988893003e-06,
      "loss": 1.1082,
      "step": 10487
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.617549055905221e-06,
      "loss": 1.2876,
      "step": 10488
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.616068122917438e-06,
      "loss": 1.259,
      "step": 10489
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.614587189929657e-06,
      "loss": 1.031,
      "step": 10490
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.613106256941874e-06,
      "loss": 1.1986,
      "step": 10491
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.611625323954092e-06,
      "loss": 1.2735,
      "step": 10492
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.610144390966309e-06,
      "loss": 1.2388,
      "step": 10493
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.608663457978527e-06,
      "loss": 1.1196,
      "step": 10494
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.607182524990745e-06,
      "loss": 1.2289,
      "step": 10495
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.6057015920029626e-06,
      "loss": 1.3321,
      "step": 10496
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.60422065901518e-06,
      "loss": 1.1326,
      "step": 10497
    },
    {
      "epoch": 2.31,
      "learning_rate": 4.602739726027398e-06,
      "loss": 1.1828,
      "step": 10498
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.601258793039615e-06,
      "loss": 1.2902,
      "step": 10499
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.599777860051833e-06,
      "loss": 0.9792,
      "step": 10500
    },
    {
      "epoch": 2.32,
      "eval_loss": NaN,
      "eval_runtime": 4.38,
      "eval_samples_per_second": 352.508,
      "eval_steps_per_second": 44.064,
      "step": 10500
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.598296927064051e-06,
      "loss": 1.3062,
      "step": 10501
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.5968159940762685e-06,
      "loss": 1.279,
      "step": 10502
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.595335061088486e-06,
      "loss": 1.2005,
      "step": 10503
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.5938541281007035e-06,
      "loss": 1.2766,
      "step": 10504
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.592373195112921e-06,
      "loss": 1.294,
      "step": 10505
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.590892262125139e-06,
      "loss": 1.386,
      "step": 10506
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.589411329137357e-06,
      "loss": 1.0247,
      "step": 10507
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.5879303961495745e-06,
      "loss": 1.2758,
      "step": 10508
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.586449463161792e-06,
      "loss": 1.2737,
      "step": 10509
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.5849685301740095e-06,
      "loss": 1.202,
      "step": 10510
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.583487597186228e-06,
      "loss": 1.1459,
      "step": 10511
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.582006664198445e-06,
      "loss": 1.2221,
      "step": 10512
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.580525731210663e-06,
      "loss": 1.3351,
      "step": 10513
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.579044798222881e-06,
      "loss": 1.2284,
      "step": 10514
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.577563865235099e-06,
      "loss": 1.3517,
      "step": 10515
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.576082932247316e-06,
      "loss": 1.3111,
      "step": 10516
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.574601999259534e-06,
      "loss": 1.2571,
      "step": 10517
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.573121066271752e-06,
      "loss": 1.1377,
      "step": 10518
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.57164013328397e-06,
      "loss": 1.1368,
      "step": 10519
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.570159200296187e-06,
      "loss": 1.3451,
      "step": 10520
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.568678267308405e-06,
      "loss": 1.0192,
      "step": 10521
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.567197334320622e-06,
      "loss": 1.2528,
      "step": 10522
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.565716401332841e-06,
      "loss": 1.2765,
      "step": 10523
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.564235468345058e-06,
      "loss": 1.5972,
      "step": 10524
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.562754535357276e-06,
      "loss": 1.2135,
      "step": 10525
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.561273602369493e-06,
      "loss": 1.1872,
      "step": 10526
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.559792669381711e-06,
      "loss": 1.2857,
      "step": 10527
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.558311736393928e-06,
      "loss": 1.1623,
      "step": 10528
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.5568308034061466e-06,
      "loss": 1.3257,
      "step": 10529
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.555349870418364e-06,
      "loss": 1.3484,
      "step": 10530
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.553868937430582e-06,
      "loss": 1.3496,
      "step": 10531
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.552388004442799e-06,
      "loss": 1.334,
      "step": 10532
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.550907071455017e-06,
      "loss": 1.386,
      "step": 10533
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.549426138467234e-06,
      "loss": 1.3673,
      "step": 10534
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.5479452054794525e-06,
      "loss": 1.4862,
      "step": 10535
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.54646427249167e-06,
      "loss": 1.2024,
      "step": 10536
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.5449833395038875e-06,
      "loss": 1.1835,
      "step": 10537
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.543502406516105e-06,
      "loss": 1.3431,
      "step": 10538
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.542021473528323e-06,
      "loss": 1.3172,
      "step": 10539
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.540540540540541e-06,
      "loss": 1.1192,
      "step": 10540
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.5390596075527584e-06,
      "loss": 1.1842,
      "step": 10541
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.537578674564976e-06,
      "loss": 1.1579,
      "step": 10542
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.536097741577194e-06,
      "loss": 1.123,
      "step": 10543
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.534616808589412e-06,
      "loss": 1.2921,
      "step": 10544
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.533135875601629e-06,
      "loss": 1.2619,
      "step": 10545
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.531654942613848e-06,
      "loss": 1.1566,
      "step": 10546
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.530174009626065e-06,
      "loss": 1.3298,
      "step": 10547
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.528693076638283e-06,
      "loss": 1.2092,
      "step": 10548
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.5272121436505e-06,
      "loss": 1.0924,
      "step": 10549
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.525731210662718e-06,
      "loss": 1.2964,
      "step": 10550
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.524250277674936e-06,
      "loss": 1.1605,
      "step": 10551
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.522769344687154e-06,
      "loss": 1.231,
      "step": 10552
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.521288411699371e-06,
      "loss": 1.1578,
      "step": 10553
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.519807478711589e-06,
      "loss": 1.2649,
      "step": 10554
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.518326545723806e-06,
      "loss": 1.1484,
      "step": 10555
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.516845612736024e-06,
      "loss": 1.1781,
      "step": 10556
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.515364679748242e-06,
      "loss": 1.1686,
      "step": 10557
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.51388374676046e-06,
      "loss": 1.0988,
      "step": 10558
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.512402813772677e-06,
      "loss": 1.2466,
      "step": 10559
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.510921880784895e-06,
      "loss": 1.1686,
      "step": 10560
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.509440947797112e-06,
      "loss": 0.9146,
      "step": 10561
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.50796001480933e-06,
      "loss": 1.2873,
      "step": 10562
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.506479081821548e-06,
      "loss": 1.1378,
      "step": 10563
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.5049981488337656e-06,
      "loss": 1.2387,
      "step": 10564
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.503517215845983e-06,
      "loss": 1.3681,
      "step": 10565
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.502036282858201e-06,
      "loss": 1.2724,
      "step": 10566
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.500555349870418e-06,
      "loss": 1.1831,
      "step": 10567
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.4990744168826365e-06,
      "loss": 1.3757,
      "step": 10568
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.497593483894854e-06,
      "loss": 1.0482,
      "step": 10569
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.4961125509070715e-06,
      "loss": 1.2707,
      "step": 10570
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.49463161791929e-06,
      "loss": 1.4799,
      "step": 10571
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.493150684931507e-06,
      "loss": 1.2261,
      "step": 10572
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.491669751943725e-06,
      "loss": 1.2977,
      "step": 10573
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.4901888189559424e-06,
      "loss": 1.0724,
      "step": 10574
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.488707885968161e-06,
      "loss": 1.2708,
      "step": 10575
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.487226952980378e-06,
      "loss": 1.173,
      "step": 10576
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.485746019992596e-06,
      "loss": 1.1118,
      "step": 10577
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.484265087004813e-06,
      "loss": 1.3592,
      "step": 10578
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.482784154017031e-06,
      "loss": 1.1366,
      "step": 10579
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.481303221029249e-06,
      "loss": 1.3944,
      "step": 10580
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.479822288041467e-06,
      "loss": 1.3451,
      "step": 10581
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.478341355053684e-06,
      "loss": 1.0201,
      "step": 10582
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.476860422065902e-06,
      "loss": 1.0966,
      "step": 10583
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.475379489078119e-06,
      "loss": 1.1305,
      "step": 10584
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.473898556090338e-06,
      "loss": 1.3045,
      "step": 10585
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.472417623102555e-06,
      "loss": 1.1357,
      "step": 10586
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.470936690114773e-06,
      "loss": 1.1973,
      "step": 10587
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.46945575712699e-06,
      "loss": 1.3554,
      "step": 10588
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.467974824139208e-06,
      "loss": 1.3412,
      "step": 10589
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.466493891151425e-06,
      "loss": 1.252,
      "step": 10590
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.465012958163644e-06,
      "loss": 1.1039,
      "step": 10591
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.463532025175861e-06,
      "loss": 1.1997,
      "step": 10592
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.462051092188079e-06,
      "loss": 1.1615,
      "step": 10593
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.460570159200296e-06,
      "loss": 1.0657,
      "step": 10594
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.459089226212514e-06,
      "loss": 1.2352,
      "step": 10595
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.457608293224732e-06,
      "loss": 1.1525,
      "step": 10596
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.4561273602369496e-06,
      "loss": 1.0563,
      "step": 10597
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.454646427249167e-06,
      "loss": 1.081,
      "step": 10598
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.453165494261385e-06,
      "loss": 1.2751,
      "step": 10599
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.451684561273603e-06,
      "loss": 1.1372,
      "step": 10600
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.4502036282858205e-06,
      "loss": 1.0629,
      "step": 10601
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.448722695298038e-06,
      "loss": 1.1231,
      "step": 10602
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.447241762310256e-06,
      "loss": 1.2815,
      "step": 10603
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.445760829322474e-06,
      "loss": 1.4105,
      "step": 10604
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.444279896334691e-06,
      "loss": 1.2622,
      "step": 10605
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.442798963346909e-06,
      "loss": 0.9151,
      "step": 10606
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.441318030359126e-06,
      "loss": 1.1349,
      "step": 10607
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.439837097371345e-06,
      "loss": 1.2745,
      "step": 10608
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.438356164383562e-06,
      "loss": 1.17,
      "step": 10609
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.43687523139578e-06,
      "loss": 0.8556,
      "step": 10610
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.435394298407997e-06,
      "loss": 1.2277,
      "step": 10611
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.433913365420215e-06,
      "loss": 1.3503,
      "step": 10612
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.432432432432433e-06,
      "loss": 1.0702,
      "step": 10613
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.430951499444651e-06,
      "loss": 1.3293,
      "step": 10614
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.429470566456868e-06,
      "loss": 1.2691,
      "step": 10615
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.427989633469086e-06,
      "loss": 0.9446,
      "step": 10616
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.426508700481303e-06,
      "loss": 1.0991,
      "step": 10617
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.425027767493521e-06,
      "loss": 1.3441,
      "step": 10618
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.423546834505739e-06,
      "loss": 1.1885,
      "step": 10619
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.422065901517957e-06,
      "loss": 1.201,
      "step": 10620
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.420584968530174e-06,
      "loss": 1.1654,
      "step": 10621
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.419104035542392e-06,
      "loss": 1.2056,
      "step": 10622
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.417623102554609e-06,
      "loss": 1.1142,
      "step": 10623
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.416142169566828e-06,
      "loss": 1.1541,
      "step": 10624
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.414661236579045e-06,
      "loss": 1.1461,
      "step": 10625
    },
    {
      "epoch": 2.34,
      "eval_loss": NaN,
      "eval_runtime": 4.373,
      "eval_samples_per_second": 353.079,
      "eval_steps_per_second": 44.135,
      "step": 10625
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.413180303591263e-06,
      "loss": 1.1551,
      "step": 10626
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.41169937060348e-06,
      "loss": 1.1847,
      "step": 10627
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.4102184376156985e-06,
      "loss": 1.1202,
      "step": 10628
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.408737504627916e-06,
      "loss": 1.2334,
      "step": 10629
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.4072565716401335e-06,
      "loss": 1.1294,
      "step": 10630
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.405775638652351e-06,
      "loss": 1.2241,
      "step": 10631
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.4042947056645694e-06,
      "loss": 0.9941,
      "step": 10632
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.402813772676787e-06,
      "loss": 1.26,
      "step": 10633
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.4013328396890045e-06,
      "loss": 1.1739,
      "step": 10634
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.399851906701222e-06,
      "loss": 1.0967,
      "step": 10635
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.39837097371344e-06,
      "loss": 1.2457,
      "step": 10636
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.396890040725658e-06,
      "loss": 1.3539,
      "step": 10637
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.395409107737875e-06,
      "loss": 1.2522,
      "step": 10638
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.393928174750093e-06,
      "loss": 1.1298,
      "step": 10639
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.39244724176231e-06,
      "loss": 1.2305,
      "step": 10640
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.390966308774529e-06,
      "loss": 1.0675,
      "step": 10641
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.389485375786746e-06,
      "loss": 1.101,
      "step": 10642
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.388004442798964e-06,
      "loss": 1.1104,
      "step": 10643
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.386523509811181e-06,
      "loss": 1.1003,
      "step": 10644
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.385042576823399e-06,
      "loss": 1.2306,
      "step": 10645
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.383561643835616e-06,
      "loss": 1.1644,
      "step": 10646
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.382080710847835e-06,
      "loss": 1.4115,
      "step": 10647
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.380599777860052e-06,
      "loss": 1.1935,
      "step": 10648
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.37911884487227e-06,
      "loss": 1.1235,
      "step": 10649
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.377637911884487e-06,
      "loss": 1.2815,
      "step": 10650
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.376156978896705e-06,
      "loss": 1.1735,
      "step": 10651
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.374676045908923e-06,
      "loss": 1.0938,
      "step": 10652
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.373195112921141e-06,
      "loss": 1.1659,
      "step": 10653
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.371714179933358e-06,
      "loss": 1.22,
      "step": 10654
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.370233246945576e-06,
      "loss": 1.3929,
      "step": 10655
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.368752313957794e-06,
      "loss": 1.2729,
      "step": 10656
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.367271380970012e-06,
      "loss": 1.2462,
      "step": 10657
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.365790447982229e-06,
      "loss": 1.2723,
      "step": 10658
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.364309514994447e-06,
      "loss": 1.2312,
      "step": 10659
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.362828582006665e-06,
      "loss": 1.2558,
      "step": 10660
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.3613476490188825e-06,
      "loss": 1.2191,
      "step": 10661
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.3598667160311e-06,
      "loss": 1.3026,
      "step": 10662
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.3583857830433175e-06,
      "loss": 0.9921,
      "step": 10663
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.356904850055536e-06,
      "loss": 1.3315,
      "step": 10664
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.355423917067753e-06,
      "loss": 1.236,
      "step": 10665
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.353942984079971e-06,
      "loss": 1.3198,
      "step": 10666
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.3524620510921884e-06,
      "loss": 1.1148,
      "step": 10667
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.350981118104406e-06,
      "loss": 1.1389,
      "step": 10668
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.349500185116624e-06,
      "loss": 1.2493,
      "step": 10669
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.348019252128842e-06,
      "loss": 1.1443,
      "step": 10670
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.346538319141059e-06,
      "loss": 1.369,
      "step": 10671
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.345057386153277e-06,
      "loss": 1.2325,
      "step": 10672
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.343576453165494e-06,
      "loss": 1.2138,
      "step": 10673
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.342095520177712e-06,
      "loss": 1.2081,
      "step": 10674
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.34061458718993e-06,
      "loss": 1.2121,
      "step": 10675
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.339133654202148e-06,
      "loss": 1.4112,
      "step": 10676
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.337652721214365e-06,
      "loss": 1.2508,
      "step": 10677
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.336171788226583e-06,
      "loss": 1.0774,
      "step": 10678
    },
    {
      "epoch": 2.35,
      "learning_rate": 4.3346908552388e-06,
      "loss": 1.0863,
      "step": 10679
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.333209922251018e-06,
      "loss": 1.1296,
      "step": 10680
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.331728989263236e-06,
      "loss": 1.0595,
      "step": 10681
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.330248056275454e-06,
      "loss": 1.2171,
      "step": 10682
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.328767123287671e-06,
      "loss": 1.0833,
      "step": 10683
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.327286190299889e-06,
      "loss": 1.2852,
      "step": 10684
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.325805257312107e-06,
      "loss": 1.1691,
      "step": 10685
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.324324324324325e-06,
      "loss": 1.027,
      "step": 10686
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.322843391336542e-06,
      "loss": 1.2641,
      "step": 10687
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.3213624583487605e-06,
      "loss": 1.0508,
      "step": 10688
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.319881525360978e-06,
      "loss": 1.2525,
      "step": 10689
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.3184005923731956e-06,
      "loss": 1.1078,
      "step": 10690
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.316919659385413e-06,
      "loss": 0.9473,
      "step": 10691
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.3154387263976315e-06,
      "loss": 1.2606,
      "step": 10692
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.313957793409849e-06,
      "loss": 1.1835,
      "step": 10693
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.3124768604220665e-06,
      "loss": 1.0547,
      "step": 10694
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.310995927434284e-06,
      "loss": 1.4976,
      "step": 10695
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.3095149944465015e-06,
      "loss": 1.1759,
      "step": 10696
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.30803406145872e-06,
      "loss": 1.1142,
      "step": 10697
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.306553128470937e-06,
      "loss": 1.059,
      "step": 10698
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.305072195483155e-06,
      "loss": 1.3235,
      "step": 10699
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.3035912624953724e-06,
      "loss": 0.9812,
      "step": 10700
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.30211032950759e-06,
      "loss": 1.1952,
      "step": 10701
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.3006293965198075e-06,
      "loss": 1.1047,
      "step": 10702
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.299148463532026e-06,
      "loss": 1.2905,
      "step": 10703
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.297667530544243e-06,
      "loss": 1.1386,
      "step": 10704
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.296186597556461e-06,
      "loss": 1.051,
      "step": 10705
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.294705664568678e-06,
      "loss": 1.2417,
      "step": 10706
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.293224731580896e-06,
      "loss": 1.2572,
      "step": 10707
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.291743798593113e-06,
      "loss": 1.0966,
      "step": 10708
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.290262865605332e-06,
      "loss": 1.2591,
      "step": 10709
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.288781932617549e-06,
      "loss": 0.979,
      "step": 10710
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.287300999629767e-06,
      "loss": 1.2965,
      "step": 10711
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.285820066641984e-06,
      "loss": 1.2631,
      "step": 10712
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.284339133654203e-06,
      "loss": 1.379,
      "step": 10713
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.28285820066642e-06,
      "loss": 1.1372,
      "step": 10714
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.281377267678638e-06,
      "loss": 1.0906,
      "step": 10715
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.279896334690855e-06,
      "loss": 1.1822,
      "step": 10716
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.278415401703074e-06,
      "loss": 1.3061,
      "step": 10717
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.276934468715291e-06,
      "loss": 1.2534,
      "step": 10718
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.275453535727509e-06,
      "loss": 1.1718,
      "step": 10719
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.273972602739727e-06,
      "loss": 1.189,
      "step": 10720
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.2724916697519445e-06,
      "loss": 1.1842,
      "step": 10721
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.271010736764162e-06,
      "loss": 1.3592,
      "step": 10722
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.2695298037763796e-06,
      "loss": 1.1199,
      "step": 10723
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.268048870788597e-06,
      "loss": 1.1506,
      "step": 10724
    },
    {
      "epoch": 2.36,
      "learning_rate": 4.2665679378008154e-06,
      "loss": 1.3266,
      "step": 10725
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.265087004813033e-06,
      "loss": 1.2912,
      "step": 10726
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.2636060718252505e-06,
      "loss": 1.1707,
      "step": 10727
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.262125138837468e-06,
      "loss": 1.2166,
      "step": 10728
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.2606442058496855e-06,
      "loss": 1.3801,
      "step": 10729
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.259163272861903e-06,
      "loss": 1.1922,
      "step": 10730
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.257682339874121e-06,
      "loss": 1.2522,
      "step": 10731
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.256201406886339e-06,
      "loss": 1.2802,
      "step": 10732
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.254720473898556e-06,
      "loss": 1.1198,
      "step": 10733
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.253239540910774e-06,
      "loss": 1.0849,
      "step": 10734
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.2517586079229915e-06,
      "loss": 1.2333,
      "step": 10735
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.250277674935209e-06,
      "loss": 1.0385,
      "step": 10736
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.248796741947427e-06,
      "loss": 1.2931,
      "step": 10737
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.247315808959645e-06,
      "loss": 1.099,
      "step": 10738
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.245834875971862e-06,
      "loss": 1.2895,
      "step": 10739
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.24435394298408e-06,
      "loss": 1.2181,
      "step": 10740
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.242873009996297e-06,
      "loss": 1.156,
      "step": 10741
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.241392077008516e-06,
      "loss": 1.0462,
      "step": 10742
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.239911144020733e-06,
      "loss": 1.2463,
      "step": 10743
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.238430211032951e-06,
      "loss": 1.0859,
      "step": 10744
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.236949278045169e-06,
      "loss": 1.1308,
      "step": 10745
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.235468345057387e-06,
      "loss": 1.4749,
      "step": 10746
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.233987412069604e-06,
      "loss": 1.1431,
      "step": 10747
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.232506479081822e-06,
      "loss": 1.2072,
      "step": 10748
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.23102554609404e-06,
      "loss": 1.2876,
      "step": 10749
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.229544613106258e-06,
      "loss": 1.2197,
      "step": 10750
    },
    {
      "epoch": 2.37,
      "eval_loss": NaN,
      "eval_runtime": 4.3795,
      "eval_samples_per_second": 352.555,
      "eval_steps_per_second": 44.069,
      "step": 10750
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.228063680118475e-06,
      "loss": 1.039,
      "step": 10751
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.226582747130693e-06,
      "loss": 1.2404,
      "step": 10752
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.225101814142911e-06,
      "loss": 1.2487,
      "step": 10753
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.2236208811551285e-06,
      "loss": 1.0024,
      "step": 10754
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.222139948167346e-06,
      "loss": 1.2588,
      "step": 10755
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.2206590151795635e-06,
      "loss": 0.9585,
      "step": 10756
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.219178082191781e-06,
      "loss": 1.0507,
      "step": 10757
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.217697149203999e-06,
      "loss": 1.0871,
      "step": 10758
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.216216216216217e-06,
      "loss": 1.1691,
      "step": 10759
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.2147352832284345e-06,
      "loss": 1.1381,
      "step": 10760
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.213254350240652e-06,
      "loss": 1.1191,
      "step": 10761
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.2117734172528695e-06,
      "loss": 1.1912,
      "step": 10762
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.210292484265087e-06,
      "loss": 1.1619,
      "step": 10763
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.2088115512773045e-06,
      "loss": 1.341,
      "step": 10764
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.207330618289523e-06,
      "loss": 1.0121,
      "step": 10765
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.20584968530174e-06,
      "loss": 1.2327,
      "step": 10766
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.204368752313958e-06,
      "loss": 0.9718,
      "step": 10767
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.2028878193261754e-06,
      "loss": 1.426,
      "step": 10768
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.201406886338393e-06,
      "loss": 1.1096,
      "step": 10769
    },
    {
      "epoch": 2.37,
      "learning_rate": 4.199925953350611e-06,
      "loss": 1.1332,
      "step": 10770
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.198445020362829e-06,
      "loss": 1.215,
      "step": 10771
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.196964087375046e-06,
      "loss": 1.2161,
      "step": 10772
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.195483154387264e-06,
      "loss": 1.3131,
      "step": 10773
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.194002221399482e-06,
      "loss": 1.2109,
      "step": 10774
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1925212884117e-06,
      "loss": 1.1224,
      "step": 10775
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.191040355423917e-06,
      "loss": 1.3358,
      "step": 10776
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.189559422436136e-06,
      "loss": 1.2703,
      "step": 10777
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.188078489448353e-06,
      "loss": 1.1017,
      "step": 10778
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.186597556460571e-06,
      "loss": 1.3682,
      "step": 10779
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.185116623472788e-06,
      "loss": 1.0351,
      "step": 10780
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.183635690485006e-06,
      "loss": 1.1339,
      "step": 10781
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.182154757497224e-06,
      "loss": 1.2061,
      "step": 10782
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.180673824509442e-06,
      "loss": 1.1594,
      "step": 10783
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.179192891521659e-06,
      "loss": 1.0931,
      "step": 10784
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.177711958533877e-06,
      "loss": 1.3502,
      "step": 10785
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.176231025546094e-06,
      "loss": 1.0469,
      "step": 10786
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1747500925583125e-06,
      "loss": 1.2653,
      "step": 10787
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.17326915957053e-06,
      "loss": 1.1379,
      "step": 10788
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1717882265827475e-06,
      "loss": 1.4332,
      "step": 10789
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.170307293594965e-06,
      "loss": 1.1852,
      "step": 10790
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1688263606071826e-06,
      "loss": 1.1597,
      "step": 10791
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1673454276194e-06,
      "loss": 1.0545,
      "step": 10792
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1658644946316184e-06,
      "loss": 1.4196,
      "step": 10793
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.164383561643836e-06,
      "loss": 1.1476,
      "step": 10794
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1629026286560535e-06,
      "loss": 1.3794,
      "step": 10795
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.161421695668271e-06,
      "loss": 1.2206,
      "step": 10796
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1599407626804885e-06,
      "loss": 1.3277,
      "step": 10797
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.158459829692707e-06,
      "loss": 1.1845,
      "step": 10798
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.156978896704924e-06,
      "loss": 1.3373,
      "step": 10799
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.155497963717142e-06,
      "loss": 1.0412,
      "step": 10800
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.1540170307293594e-06,
      "loss": 1.372,
      "step": 10801
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.152536097741578e-06,
      "loss": 1.1336,
      "step": 10802
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.151055164753795e-06,
      "loss": 1.1569,
      "step": 10803
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.149574231766013e-06,
      "loss": 1.1661,
      "step": 10804
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.14809329877823e-06,
      "loss": 1.2392,
      "step": 10805
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.146612365790449e-06,
      "loss": 1.2493,
      "step": 10806
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.145131432802666e-06,
      "loss": 1.5086,
      "step": 10807
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.143650499814884e-06,
      "loss": 1.0502,
      "step": 10808
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.142169566827101e-06,
      "loss": 1.142,
      "step": 10809
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.14068863383932e-06,
      "loss": 1.3176,
      "step": 10810
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.139207700851537e-06,
      "loss": 0.9337,
      "step": 10811
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.137726767863755e-06,
      "loss": 1.2936,
      "step": 10812
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.136245834875972e-06,
      "loss": 1.1593,
      "step": 10813
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.13476490188819e-06,
      "loss": 1.2894,
      "step": 10814
    },
    {
      "epoch": 2.38,
      "learning_rate": 4.133283968900408e-06,
      "loss": 1.324,
      "step": 10815
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.1318030359126256e-06,
      "loss": 1.3121,
      "step": 10816
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.130322102924843e-06,
      "loss": 1.0271,
      "step": 10817
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.128841169937061e-06,
      "loss": 1.4595,
      "step": 10818
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.127360236949278e-06,
      "loss": 1.3174,
      "step": 10819
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.125879303961496e-06,
      "loss": 1.1012,
      "step": 10820
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.124398370973714e-06,
      "loss": 1.0117,
      "step": 10821
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.1229174379859315e-06,
      "loss": 1.1759,
      "step": 10822
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.121436504998149e-06,
      "loss": 1.1143,
      "step": 10823
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.1199555720103666e-06,
      "loss": 1.1244,
      "step": 10824
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.118474639022584e-06,
      "loss": 1.1602,
      "step": 10825
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.1169937060348024e-06,
      "loss": 1.1958,
      "step": 10826
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.11551277304702e-06,
      "loss": 1.2186,
      "step": 10827
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.1140318400592375e-06,
      "loss": 1.1758,
      "step": 10828
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.112550907071455e-06,
      "loss": 1.1293,
      "step": 10829
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.1110699740836725e-06,
      "loss": 1.1915,
      "step": 10830
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.109589041095891e-06,
      "loss": 1.3622,
      "step": 10831
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.108108108108108e-06,
      "loss": 1.2189,
      "step": 10832
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.106627175120326e-06,
      "loss": 1.4211,
      "step": 10833
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.105146242132544e-06,
      "loss": 1.1891,
      "step": 10834
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.103665309144762e-06,
      "loss": 1.256,
      "step": 10835
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.102184376156979e-06,
      "loss": 1.0303,
      "step": 10836
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.100703443169197e-06,
      "loss": 1.3435,
      "step": 10837
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.099222510181415e-06,
      "loss": 1.1092,
      "step": 10838
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.097741577193633e-06,
      "loss": 1.0441,
      "step": 10839
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.09626064420585e-06,
      "loss": 1.2217,
      "step": 10840
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.094779711218068e-06,
      "loss": 1.3368,
      "step": 10841
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.093298778230285e-06,
      "loss": 1.1272,
      "step": 10842
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.091817845242504e-06,
      "loss": 1.3277,
      "step": 10843
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.090336912254721e-06,
      "loss": 1.0338,
      "step": 10844
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.088855979266939e-06,
      "loss": 1.0971,
      "step": 10845
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.087375046279156e-06,
      "loss": 1.0504,
      "step": 10846
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.085894113291374e-06,
      "loss": 1.0903,
      "step": 10847
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.084413180303591e-06,
      "loss": 1.1802,
      "step": 10848
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.0829322473158096e-06,
      "loss": 0.9902,
      "step": 10849
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.081451314328027e-06,
      "loss": 1.3937,
      "step": 10850
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.079970381340245e-06,
      "loss": 1.1861,
      "step": 10851
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.078489448352462e-06,
      "loss": 0.9987,
      "step": 10852
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.07700851536468e-06,
      "loss": 1.3981,
      "step": 10853
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.075527582376898e-06,
      "loss": 1.3135,
      "step": 10854
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.0740466493891155e-06,
      "loss": 1.1793,
      "step": 10855
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.072565716401333e-06,
      "loss": 1.204,
      "step": 10856
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.0710847834135505e-06,
      "loss": 1.3375,
      "step": 10857
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.069603850425768e-06,
      "loss": 1.386,
      "step": 10858
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.068122917437986e-06,
      "loss": 1.3071,
      "step": 10859
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.066641984450204e-06,
      "loss": 1.2302,
      "step": 10860
    },
    {
      "epoch": 2.39,
      "learning_rate": 4.0651610514624215e-06,
      "loss": 1.0292,
      "step": 10861
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.063680118474639e-06,
      "loss": 1.4278,
      "step": 10862
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.062199185486857e-06,
      "loss": 1.2208,
      "step": 10863
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.060718252499075e-06,
      "loss": 1.2192,
      "step": 10864
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.059237319511292e-06,
      "loss": 1.0741,
      "step": 10865
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.057756386523511e-06,
      "loss": 1.2573,
      "step": 10866
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.056275453535728e-06,
      "loss": 1.2904,
      "step": 10867
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.054794520547946e-06,
      "loss": 1.2995,
      "step": 10868
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.053313587560163e-06,
      "loss": 1.0205,
      "step": 10869
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.051832654572381e-06,
      "loss": 1.1689,
      "step": 10870
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.050351721584599e-06,
      "loss": 1.3707,
      "step": 10871
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.048870788596817e-06,
      "loss": 1.3165,
      "step": 10872
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.047389855609034e-06,
      "loss": 1.0974,
      "step": 10873
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.045908922621252e-06,
      "loss": 1.3279,
      "step": 10874
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.044427989633469e-06,
      "loss": 1.1753,
      "step": 10875
    },
    {
      "epoch": 2.4,
      "eval_loss": NaN,
      "eval_runtime": 4.3775,
      "eval_samples_per_second": 352.714,
      "eval_steps_per_second": 44.089,
      "step": 10875
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.042947056645687e-06,
      "loss": 1.0607,
      "step": 10876
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.041466123657905e-06,
      "loss": 1.4564,
      "step": 10877
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.039985190670123e-06,
      "loss": 1.2909,
      "step": 10878
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.03850425768234e-06,
      "loss": 1.1481,
      "step": 10879
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.037023324694558e-06,
      "loss": 1.3327,
      "step": 10880
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.035542391706775e-06,
      "loss": 1.3951,
      "step": 10881
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.0340614587189935e-06,
      "loss": 1.447,
      "step": 10882
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.032580525731211e-06,
      "loss": 1.1887,
      "step": 10883
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.0310995927434286e-06,
      "loss": 0.9784,
      "step": 10884
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.029618659755646e-06,
      "loss": 1.1006,
      "step": 10885
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.028137726767864e-06,
      "loss": 1.2266,
      "step": 10886
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.026656793780082e-06,
      "loss": 1.1909,
      "step": 10887
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.0251758607922995e-06,
      "loss": 1.2066,
      "step": 10888
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.023694927804517e-06,
      "loss": 1.2525,
      "step": 10889
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.0222139948167345e-06,
      "loss": 1.1704,
      "step": 10890
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.020733061828953e-06,
      "loss": 1.1571,
      "step": 10891
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.01925212884117e-06,
      "loss": 1.1641,
      "step": 10892
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.017771195853388e-06,
      "loss": 1.4518,
      "step": 10893
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.0162902628656054e-06,
      "loss": 1.2741,
      "step": 10894
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.014809329877824e-06,
      "loss": 1.1406,
      "step": 10895
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.013328396890041e-06,
      "loss": 1.2353,
      "step": 10896
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.011847463902259e-06,
      "loss": 1.3226,
      "step": 10897
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.010366530914476e-06,
      "loss": 1.1883,
      "step": 10898
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.008885597926695e-06,
      "loss": 1.2958,
      "step": 10899
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.007404664938912e-06,
      "loss": 1.2537,
      "step": 10900
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.00592373195113e-06,
      "loss": 1.1784,
      "step": 10901
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.004442798963347e-06,
      "loss": 1.235,
      "step": 10902
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.002961865975565e-06,
      "loss": 1.0901,
      "step": 10903
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.001480932987782e-06,
      "loss": 1.1945,
      "step": 10904
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.0164,
      "step": 10905
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.998519067012218e-06,
      "loss": 1.4534,
      "step": 10906
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.997038134024436e-06,
      "loss": 1.0977,
      "step": 10907
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.995557201036653e-06,
      "loss": 1.1002,
      "step": 10908
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.994076268048871e-06,
      "loss": 1.4249,
      "step": 10909
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.992595335061088e-06,
      "loss": 1.3321,
      "step": 10910
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.991114402073307e-06,
      "loss": 1.301,
      "step": 10911
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.989633469085524e-06,
      "loss": 1.1251,
      "step": 10912
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.988152536097742e-06,
      "loss": 1.1993,
      "step": 10913
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.986671603109959e-06,
      "loss": 1.3094,
      "step": 10914
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.985190670122177e-06,
      "loss": 1.3656,
      "step": 10915
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.983709737134395e-06,
      "loss": 1.2681,
      "step": 10916
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.9822288041466126e-06,
      "loss": 1.2763,
      "step": 10917
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.98074787115883e-06,
      "loss": 1.0729,
      "step": 10918
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.9792669381710484e-06,
      "loss": 1.162,
      "step": 10919
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.977786005183266e-06,
      "loss": 1.2792,
      "step": 10920
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.9763050721954835e-06,
      "loss": 1.1413,
      "step": 10921
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.974824139207701e-06,
      "loss": 1.2299,
      "step": 10922
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.973343206219919e-06,
      "loss": 1.1552,
      "step": 10923
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.971862273232137e-06,
      "loss": 1.0354,
      "step": 10924
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.970381340244354e-06,
      "loss": 1.2756,
      "step": 10925
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.968900407256572e-06,
      "loss": 1.3135,
      "step": 10926
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.96741947426879e-06,
      "loss": 1.1872,
      "step": 10927
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.965938541281008e-06,
      "loss": 1.1155,
      "step": 10928
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.964457608293225e-06,
      "loss": 1.1542,
      "step": 10929
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.962976675305443e-06,
      "loss": 1.1028,
      "step": 10930
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.96149574231766e-06,
      "loss": 1.1126,
      "step": 10931
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.960014809329878e-06,
      "loss": 1.2155,
      "step": 10932
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.958533876342096e-06,
      "loss": 1.1528,
      "step": 10933
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.957052943354314e-06,
      "loss": 1.6005,
      "step": 10934
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.955572010366531e-06,
      "loss": 1.1444,
      "step": 10935
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.954091077378749e-06,
      "loss": 1.1445,
      "step": 10936
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.952610144390966e-06,
      "loss": 1.3163,
      "step": 10937
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.951129211403184e-06,
      "loss": 1.2872,
      "step": 10938
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.949648278415402e-06,
      "loss": 1.4395,
      "step": 10939
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.94816734542762e-06,
      "loss": 1.4065,
      "step": 10940
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.946686412439837e-06,
      "loss": 1.2046,
      "step": 10941
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.945205479452055e-06,
      "loss": 1.0822,
      "step": 10942
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.943724546464272e-06,
      "loss": 1.1752,
      "step": 10943
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.942243613476491e-06,
      "loss": 1.1714,
      "step": 10944
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.940762680488708e-06,
      "loss": 1.189,
      "step": 10945
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.939281747500926e-06,
      "loss": 1.0287,
      "step": 10946
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.937800814513143e-06,
      "loss": 1.2465,
      "step": 10947
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.9363198815253615e-06,
      "loss": 1.0906,
      "step": 10948
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.934838948537579e-06,
      "loss": 1.1437,
      "step": 10949
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.9333580155497965e-06,
      "loss": 1.2045,
      "step": 10950
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.931877082562015e-06,
      "loss": 1.1539,
      "step": 10951
    },
    {
      "epoch": 2.41,
      "learning_rate": 3.9303961495742324e-06,
      "loss": 1.305,
      "step": 10952
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.92891521658645e-06,
      "loss": 1.3957,
      "step": 10953
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.9274342835986675e-06,
      "loss": 1.2299,
      "step": 10954
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.925953350610886e-06,
      "loss": 1.1688,
      "step": 10955
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.924472417623103e-06,
      "loss": 1.21,
      "step": 10956
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.922991484635321e-06,
      "loss": 1.2917,
      "step": 10957
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.921510551647538e-06,
      "loss": 1.2575,
      "step": 10958
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.920029618659756e-06,
      "loss": 1.3764,
      "step": 10959
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.918548685671973e-06,
      "loss": 1.2158,
      "step": 10960
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.917067752684192e-06,
      "loss": 1.2326,
      "step": 10961
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.915586819696409e-06,
      "loss": 1.1374,
      "step": 10962
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.914105886708627e-06,
      "loss": 1.1786,
      "step": 10963
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.912624953720844e-06,
      "loss": 1.2039,
      "step": 10964
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.911144020733062e-06,
      "loss": 0.9455,
      "step": 10965
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.909663087745279e-06,
      "loss": 1.2134,
      "step": 10966
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.908182154757498e-06,
      "loss": 1.1398,
      "step": 10967
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.906701221769715e-06,
      "loss": 1.1778,
      "step": 10968
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.905220288781933e-06,
      "loss": 1.3067,
      "step": 10969
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.90373935579415e-06,
      "loss": 1.2997,
      "step": 10970
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.902258422806368e-06,
      "loss": 1.1899,
      "step": 10971
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.900777489818586e-06,
      "loss": 1.146,
      "step": 10972
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.899296556830804e-06,
      "loss": 1.2562,
      "step": 10973
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.897815623843021e-06,
      "loss": 1.1653,
      "step": 10974
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.896334690855239e-06,
      "loss": 1.3126,
      "step": 10975
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.894853757867457e-06,
      "loss": 1.1605,
      "step": 10976
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.893372824879675e-06,
      "loss": 1.0634,
      "step": 10977
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.891891891891892e-06,
      "loss": 1.4219,
      "step": 10978
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.89041095890411e-06,
      "loss": 1.035,
      "step": 10979
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.888930025916328e-06,
      "loss": 1.1726,
      "step": 10980
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.8874490929285455e-06,
      "loss": 1.2149,
      "step": 10981
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.885968159940763e-06,
      "loss": 1.1121,
      "step": 10982
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.884487226952981e-06,
      "loss": 1.2352,
      "step": 10983
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.883006293965199e-06,
      "loss": 1.126,
      "step": 10984
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.881525360977416e-06,
      "loss": 1.2302,
      "step": 10985
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.880044427989634e-06,
      "loss": 1.1607,
      "step": 10986
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.8785634950018515e-06,
      "loss": 1.1258,
      "step": 10987
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.877082562014069e-06,
      "loss": 1.0477,
      "step": 10988
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.875601629026287e-06,
      "loss": 1.0849,
      "step": 10989
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.874120696038505e-06,
      "loss": 1.1656,
      "step": 10990
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.872639763050722e-06,
      "loss": 1.2267,
      "step": 10991
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.87115883006294e-06,
      "loss": 1.4227,
      "step": 10992
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.869677897075157e-06,
      "loss": 1.3691,
      "step": 10993
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.868196964087375e-06,
      "loss": 1.363,
      "step": 10994
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.866716031099593e-06,
      "loss": 1.0785,
      "step": 10995
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.865235098111811e-06,
      "loss": 1.2458,
      "step": 10996
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.863754165124028e-06,
      "loss": 1.0745,
      "step": 10997
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.862273232136246e-06,
      "loss": 1.138,
      "step": 10998
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.860792299148463e-06,
      "loss": 1.3618,
      "step": 10999
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.859311366160682e-06,
      "loss": 1.1468,
      "step": 11000
    },
    {
      "epoch": 2.43,
      "eval_loss": NaN,
      "eval_runtime": 4.368,
      "eval_samples_per_second": 353.477,
      "eval_steps_per_second": 44.185,
      "step": 11000
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.857830433172899e-06,
      "loss": 1.4795,
      "step": 11001
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.856349500185117e-06,
      "loss": 1.1769,
      "step": 11002
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.854868567197334e-06,
      "loss": 1.0414,
      "step": 11003
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.853387634209552e-06,
      "loss": 1.1763,
      "step": 11004
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.85190670122177e-06,
      "loss": 1.5323,
      "step": 11005
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.850425768233988e-06,
      "loss": 1.2022,
      "step": 11006
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.848944835246205e-06,
      "loss": 1.2115,
      "step": 11007
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8474639022584235e-06,
      "loss": 1.1414,
      "step": 11008
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.845982969270641e-06,
      "loss": 1.3516,
      "step": 11009
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8445020362828586e-06,
      "loss": 1.32,
      "step": 11010
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.843021103295076e-06,
      "loss": 1.0551,
      "step": 11011
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8415401703072945e-06,
      "loss": 1.3595,
      "step": 11012
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.840059237319512e-06,
      "loss": 1.2403,
      "step": 11013
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8385783043317295e-06,
      "loss": 1.1264,
      "step": 11014
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.837097371343947e-06,
      "loss": 1.1814,
      "step": 11015
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8356164383561645e-06,
      "loss": 1.1123,
      "step": 11016
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.834135505368383e-06,
      "loss": 1.4688,
      "step": 11017
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8326545723806e-06,
      "loss": 1.3606,
      "step": 11018
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.831173639392818e-06,
      "loss": 1.2777,
      "step": 11019
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8296927064050354e-06,
      "loss": 1.1425,
      "step": 11020
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.828211773417253e-06,
      "loss": 1.085,
      "step": 11021
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8267308404294705e-06,
      "loss": 1.3055,
      "step": 11022
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.825249907441689e-06,
      "loss": 1.3342,
      "step": 11023
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.823768974453906e-06,
      "loss": 1.1557,
      "step": 11024
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.822288041466124e-06,
      "loss": 1.2129,
      "step": 11025
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.820807108478341e-06,
      "loss": 1.0721,
      "step": 11026
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.819326175490559e-06,
      "loss": 1.3932,
      "step": 11027
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.817845242502777e-06,
      "loss": 1.2884,
      "step": 11028
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.816364309514995e-06,
      "loss": 1.2068,
      "step": 11029
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.814883376527212e-06,
      "loss": 0.9715,
      "step": 11030
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8134024435394302e-06,
      "loss": 1.0571,
      "step": 11031
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8119215105516478e-06,
      "loss": 1.0822,
      "step": 11032
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8104405775638653e-06,
      "loss": 1.235,
      "step": 11033
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8089596445760836e-06,
      "loss": 1.1375,
      "step": 11034
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.807478711588301e-06,
      "loss": 1.3349,
      "step": 11035
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8059977786005187e-06,
      "loss": 1.5204,
      "step": 11036
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.804516845612736e-06,
      "loss": 1.1199,
      "step": 11037
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8030359126249537e-06,
      "loss": 1.165,
      "step": 11038
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8015549796371716e-06,
      "loss": 1.4442,
      "step": 11039
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.8000740466493896e-06,
      "loss": 1.2323,
      "step": 11040
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.798593113661607e-06,
      "loss": 1.3675,
      "step": 11041
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.7971121806738246e-06,
      "loss": 1.2062,
      "step": 11042
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7956312476860426e-06,
      "loss": 1.2172,
      "step": 11043
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.79415031469826e-06,
      "loss": 1.3381,
      "step": 11044
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.792669381710478e-06,
      "loss": 1.2368,
      "step": 11045
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.791188448722696e-06,
      "loss": 1.2834,
      "step": 11046
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7897075157349135e-06,
      "loss": 1.2955,
      "step": 11047
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.788226582747131e-06,
      "loss": 1.1069,
      "step": 11048
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7867456497593485e-06,
      "loss": 1.2883,
      "step": 11049
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.785264716771566e-06,
      "loss": 1.2437,
      "step": 11050
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7837837837837844e-06,
      "loss": 1.2507,
      "step": 11051
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.782302850796002e-06,
      "loss": 1.2182,
      "step": 11052
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7808219178082194e-06,
      "loss": 1.3691,
      "step": 11053
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.779340984820437e-06,
      "loss": 1.2905,
      "step": 11054
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.777860051832655e-06,
      "loss": 1.2614,
      "step": 11055
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.776379118844873e-06,
      "loss": 1.102,
      "step": 11056
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7748981858570903e-06,
      "loss": 1.3573,
      "step": 11057
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.773417252869308e-06,
      "loss": 1.4085,
      "step": 11058
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.771936319881526e-06,
      "loss": 1.3001,
      "step": 11059
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7704553868937433e-06,
      "loss": 1.321,
      "step": 11060
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.768974453905961e-06,
      "loss": 1.4008,
      "step": 11061
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.767493520918179e-06,
      "loss": 1.0323,
      "step": 11062
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7660125879303967e-06,
      "loss": 1.2499,
      "step": 11063
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7645316549426142e-06,
      "loss": 1.1561,
      "step": 11064
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7630507219548317e-06,
      "loss": 1.104,
      "step": 11065
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7615697889670493e-06,
      "loss": 1.2856,
      "step": 11066
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7600888559792668e-06,
      "loss": 1.2462,
      "step": 11067
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.758607922991485e-06,
      "loss": 1.2343,
      "step": 11068
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7571269900037027e-06,
      "loss": 1.1377,
      "step": 11069
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.75564605701592e-06,
      "loss": 1.2759,
      "step": 11070
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.754165124028138e-06,
      "loss": 1.0559,
      "step": 11071
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7526841910403556e-06,
      "loss": 1.2215,
      "step": 11072
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7512032580525736e-06,
      "loss": 1.4247,
      "step": 11073
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.749722325064791e-06,
      "loss": 1.2315,
      "step": 11074
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.748241392077009e-06,
      "loss": 1.0039,
      "step": 11075
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7467604590892265e-06,
      "loss": 1.2797,
      "step": 11076
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.745279526101444e-06,
      "loss": 1.2656,
      "step": 11077
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7437985931136616e-06,
      "loss": 1.1131,
      "step": 11078
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.74231766012588e-06,
      "loss": 1.0826,
      "step": 11079
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7408367271380975e-06,
      "loss": 1.2263,
      "step": 11080
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.739355794150315e-06,
      "loss": 1.3887,
      "step": 11081
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7378748611625325e-06,
      "loss": 1.0638,
      "step": 11082
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.73639392817475e-06,
      "loss": 1.3572,
      "step": 11083
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7349129951869684e-06,
      "loss": 1.3421,
      "step": 11084
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.733432062199186e-06,
      "loss": 1.2556,
      "step": 11085
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7319511292114034e-06,
      "loss": 1.1948,
      "step": 11086
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.7304701962236214e-06,
      "loss": 1.33,
      "step": 11087
    },
    {
      "epoch": 2.44,
      "learning_rate": 3.728989263235839e-06,
      "loss": 0.988,
      "step": 11088
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7275083302480564e-06,
      "loss": 1.0617,
      "step": 11089
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7260273972602743e-06,
      "loss": 1.2066,
      "step": 11090
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7245464642724923e-06,
      "loss": 1.1364,
      "step": 11091
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7230655312847098e-06,
      "loss": 1.169,
      "step": 11092
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7215845982969273e-06,
      "loss": 1.2,
      "step": 11093
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.720103665309145e-06,
      "loss": 1.1649,
      "step": 11094
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7186227323213623e-06,
      "loss": 1.2202,
      "step": 11095
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7171417993335807e-06,
      "loss": 1.0834,
      "step": 11096
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7156608663457982e-06,
      "loss": 1.2918,
      "step": 11097
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7141799333580157e-06,
      "loss": 1.2934,
      "step": 11098
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7126990003702332e-06,
      "loss": 1.4608,
      "step": 11099
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.711218067382451e-06,
      "loss": 1.1391,
      "step": 11100
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.709737134394669e-06,
      "loss": 1.2677,
      "step": 11101
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7082562014068866e-06,
      "loss": 1.1985,
      "step": 11102
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7067752684191046e-06,
      "loss": 1.1686,
      "step": 11103
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.705294335431322e-06,
      "loss": 1.2198,
      "step": 11104
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7038134024435396e-06,
      "loss": 1.2932,
      "step": 11105
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.702332469455757e-06,
      "loss": 1.3323,
      "step": 11106
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7008515364679755e-06,
      "loss": 1.098,
      "step": 11107
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.699370603480193e-06,
      "loss": 1.0665,
      "step": 11108
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6978896704924105e-06,
      "loss": 1.238,
      "step": 11109
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.696408737504628e-06,
      "loss": 1.0567,
      "step": 11110
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6949278045168456e-06,
      "loss": 1.1241,
      "step": 11111
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6934468715290635e-06,
      "loss": 1.0589,
      "step": 11112
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6919659385412814e-06,
      "loss": 1.1682,
      "step": 11113
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.690485005553499e-06,
      "loss": 1.2405,
      "step": 11114
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6890040725657165e-06,
      "loss": 1.0014,
      "step": 11115
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6875231395779344e-06,
      "loss": 1.1511,
      "step": 11116
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.686042206590152e-06,
      "loss": 1.4767,
      "step": 11117
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.68456127360237e-06,
      "loss": 1.1845,
      "step": 11118
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.683080340614588e-06,
      "loss": 1.2555,
      "step": 11119
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6815994076268053e-06,
      "loss": 1.1981,
      "step": 11120
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.680118474639023e-06,
      "loss": 1.266,
      "step": 11121
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6786375416512404e-06,
      "loss": 0.983,
      "step": 11122
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.677156608663458e-06,
      "loss": 1.0599,
      "step": 11123
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6756756756756763e-06,
      "loss": 1.3541,
      "step": 11124
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6741947426878938e-06,
      "loss": 1.1392,
      "step": 11125
    },
    {
      "epoch": 2.45,
      "eval_loss": NaN,
      "eval_runtime": 4.3691,
      "eval_samples_per_second": 353.387,
      "eval_steps_per_second": 44.173,
      "step": 11125
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6727138097001113e-06,
      "loss": 1.122,
      "step": 11126
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.671232876712329e-06,
      "loss": 1.1554,
      "step": 11127
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6697519437245467e-06,
      "loss": 1.1382,
      "step": 11128
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6682710107367647e-06,
      "loss": 1.2555,
      "step": 11129
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.666790077748982e-06,
      "loss": 1.2189,
      "step": 11130
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6653091447611997e-06,
      "loss": 1.1199,
      "step": 11131
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6638282117734177e-06,
      "loss": 1.3478,
      "step": 11132
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.662347278785635e-06,
      "loss": 1.3882,
      "step": 11133
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6608663457978527e-06,
      "loss": 0.9311,
      "step": 11134
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.659385412810071e-06,
      "loss": 1.2437,
      "step": 11135
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6579044798222886e-06,
      "loss": 1.0786,
      "step": 11136
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.656423546834506e-06,
      "loss": 1.2038,
      "step": 11137
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6549426138467236e-06,
      "loss": 1.3565,
      "step": 11138
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.653461680858941e-06,
      "loss": 1.1359,
      "step": 11139
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.651980747871159e-06,
      "loss": 1.379,
      "step": 11140
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.650499814883377e-06,
      "loss": 1.305,
      "step": 11141
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6490188818955945e-06,
      "loss": 1.0115,
      "step": 11142
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.647537948907812e-06,
      "loss": 1.1471,
      "step": 11143
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.64605701592003e-06,
      "loss": 1.096,
      "step": 11144
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6445760829322475e-06,
      "loss": 1.5371,
      "step": 11145
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6430951499444654e-06,
      "loss": 1.2668,
      "step": 11146
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.641614216956683e-06,
      "loss": 1.3176,
      "step": 11147
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.640133283968901e-06,
      "loss": 1.1569,
      "step": 11148
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6386523509811184e-06,
      "loss": 0.9117,
      "step": 11149
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.637171417993336e-06,
      "loss": 1.2555,
      "step": 11150
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6356904850055534e-06,
      "loss": 1.3062,
      "step": 11151
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.634209552017772e-06,
      "loss": 1.4079,
      "step": 11152
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6327286190299893e-06,
      "loss": 1.2497,
      "step": 11153
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.631247686042207e-06,
      "loss": 1.1404,
      "step": 11154
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6297667530544244e-06,
      "loss": 1.2071,
      "step": 11155
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6282858200666423e-06,
      "loss": 1.2731,
      "step": 11156
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6268048870788602e-06,
      "loss": 1.2552,
      "step": 11157
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6253239540910778e-06,
      "loss": 1.2065,
      "step": 11158
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6238430211032953e-06,
      "loss": 1.1605,
      "step": 11159
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6223620881155132e-06,
      "loss": 1.3796,
      "step": 11160
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6208811551277307e-06,
      "loss": 1.3757,
      "step": 11161
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6194002221399482e-06,
      "loss": 1.3245,
      "step": 11162
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.617919289152166e-06,
      "loss": 1.1033,
      "step": 11163
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.616438356164384e-06,
      "loss": 1.3061,
      "step": 11164
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6149574231766016e-06,
      "loss": 1.3197,
      "step": 11165
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.613476490188819e-06,
      "loss": 1.2299,
      "step": 11166
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6119955572010367e-06,
      "loss": 1.3293,
      "step": 11167
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.610514624213254e-06,
      "loss": 1.1305,
      "step": 11168
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6090336912254726e-06,
      "loss": 1.184,
      "step": 11169
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.60755275823769e-06,
      "loss": 1.2039,
      "step": 11170
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6060718252499076e-06,
      "loss": 1.1768,
      "step": 11171
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6045908922621255e-06,
      "loss": 1.4008,
      "step": 11172
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.603109959274343e-06,
      "loss": 1.1509,
      "step": 11173
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.601629026286561e-06,
      "loss": 1.0942,
      "step": 11174
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.6001480932987785e-06,
      "loss": 1.0486,
      "step": 11175
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.5986671603109964e-06,
      "loss": 1.1543,
      "step": 11176
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.597186227323214e-06,
      "loss": 1.2503,
      "step": 11177
    },
    {
      "epoch": 2.46,
      "learning_rate": 3.5957052943354315e-06,
      "loss": 1.2065,
      "step": 11178
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.594224361347649e-06,
      "loss": 1.4018,
      "step": 11179
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5927434283598674e-06,
      "loss": 1.1207,
      "step": 11180
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.591262495372085e-06,
      "loss": 1.2637,
      "step": 11181
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5897815623843024e-06,
      "loss": 1.3132,
      "step": 11182
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.58830062939652e-06,
      "loss": 1.0697,
      "step": 11183
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5868196964087374e-06,
      "loss": 1.2834,
      "step": 11184
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.585338763420956e-06,
      "loss": 1.1789,
      "step": 11185
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5838578304331733e-06,
      "loss": 1.1034,
      "step": 11186
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.582376897445391e-06,
      "loss": 1.0509,
      "step": 11187
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5808959644576088e-06,
      "loss": 1.2493,
      "step": 11188
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5794150314698263e-06,
      "loss": 1.3186,
      "step": 11189
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.577934098482044e-06,
      "loss": 1.1904,
      "step": 11190
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5764531654942617e-06,
      "loss": 1.2987,
      "step": 11191
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5749722325064797e-06,
      "loss": 1.1555,
      "step": 11192
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.573491299518697e-06,
      "loss": 1.1446,
      "step": 11193
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5720103665309147e-06,
      "loss": 1.0647,
      "step": 11194
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5705294335431322e-06,
      "loss": 1.307,
      "step": 11195
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5690485005553498e-06,
      "loss": 1.3839,
      "step": 11196
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.567567567567568e-06,
      "loss": 1.1413,
      "step": 11197
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5660866345797856e-06,
      "loss": 1.3915,
      "step": 11198
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.564605701592003e-06,
      "loss": 1.2932,
      "step": 11199
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5631247686042207e-06,
      "loss": 1.0552,
      "step": 11200
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5616438356164386e-06,
      "loss": 1.016,
      "step": 11201
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5601629026286565e-06,
      "loss": 1.2072,
      "step": 11202
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.558681969640874e-06,
      "loss": 1.2499,
      "step": 11203
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.557201036653092e-06,
      "loss": 1.2077,
      "step": 11204
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5557201036653095e-06,
      "loss": 1.2121,
      "step": 11205
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.554239170677527e-06,
      "loss": 1.3125,
      "step": 11206
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5527582376897446e-06,
      "loss": 1.3889,
      "step": 11207
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.551277304701963e-06,
      "loss": 1.2681,
      "step": 11208
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5497963717141804e-06,
      "loss": 1.3384,
      "step": 11209
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.548315438726398e-06,
      "loss": 1.2158,
      "step": 11210
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5468345057386155e-06,
      "loss": 1.166,
      "step": 11211
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.545353572750833e-06,
      "loss": 1.2397,
      "step": 11212
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.543872639763051e-06,
      "loss": 1.2188,
      "step": 11213
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.542391706775269e-06,
      "loss": 1.1913,
      "step": 11214
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5409107737874864e-06,
      "loss": 1.1849,
      "step": 11215
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.539429840799704e-06,
      "loss": 1.3078,
      "step": 11216
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.537948907811922e-06,
      "loss": 1.2713,
      "step": 11217
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5364679748241394e-06,
      "loss": 1.3343,
      "step": 11218
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5349870418363573e-06,
      "loss": 1.2227,
      "step": 11219
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5335061088485752e-06,
      "loss": 1.1522,
      "step": 11220
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5320251758607928e-06,
      "loss": 1.0338,
      "step": 11221
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5305442428730103e-06,
      "loss": 1.1391,
      "step": 11222
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.529063309885228e-06,
      "loss": 1.1606,
      "step": 11223
    },
    {
      "epoch": 2.47,
      "learning_rate": 3.5275823768974453e-06,
      "loss": 1.1065,
      "step": 11224
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.5261014439096637e-06,
      "loss": 1.4256,
      "step": 11225
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.524620510921881e-06,
      "loss": 1.0788,
      "step": 11226
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.5231395779340987e-06,
      "loss": 1.1625,
      "step": 11227
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.5216586449463162e-06,
      "loss": 1.2009,
      "step": 11228
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.520177711958534e-06,
      "loss": 1.0276,
      "step": 11229
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.518696778970752e-06,
      "loss": 1.2844,
      "step": 11230
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.5172158459829696e-06,
      "loss": 1.2477,
      "step": 11231
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.515734912995187e-06,
      "loss": 1.184,
      "step": 11232
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.514253980007405e-06,
      "loss": 1.1099,
      "step": 11233
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.5127730470196226e-06,
      "loss": 1.3256,
      "step": 11234
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.51129211403184e-06,
      "loss": 1.4356,
      "step": 11235
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.5098111810440585e-06,
      "loss": 1.0592,
      "step": 11236
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.508330248056276e-06,
      "loss": 1.1,
      "step": 11237
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.5068493150684935e-06,
      "loss": 1.2424,
      "step": 11238
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.505368382080711e-06,
      "loss": 1.1879,
      "step": 11239
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.5038874490929285e-06,
      "loss": 1.3221,
      "step": 11240
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.502406516105146e-06,
      "loss": 1.0257,
      "step": 11241
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.5009255831173644e-06,
      "loss": 1.4059,
      "step": 11242
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.499444650129582e-06,
      "loss": 1.2951,
      "step": 11243
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4979637171417995e-06,
      "loss": 1.2626,
      "step": 11244
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4964827841540174e-06,
      "loss": 1.2986,
      "step": 11245
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.495001851166235e-06,
      "loss": 1.3046,
      "step": 11246
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.493520918178453e-06,
      "loss": 1.1149,
      "step": 11247
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4920399851906704e-06,
      "loss": 1.1635,
      "step": 11248
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4905590522028883e-06,
      "loss": 1.5659,
      "step": 11249
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.489078119215106e-06,
      "loss": 1.29,
      "step": 11250
    },
    {
      "epoch": 2.48,
      "eval_loss": NaN,
      "eval_runtime": 4.3726,
      "eval_samples_per_second": 353.105,
      "eval_steps_per_second": 44.138,
      "step": 11250
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4875971862273233e-06,
      "loss": 1.0835,
      "step": 11251
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.486116253239541e-06,
      "loss": 1.1251,
      "step": 11252
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4846353202517592e-06,
      "loss": 1.3129,
      "step": 11253
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4831543872639767e-06,
      "loss": 1.2708,
      "step": 11254
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4816734542761943e-06,
      "loss": 1.2107,
      "step": 11255
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4801925212884118e-06,
      "loss": 1.3738,
      "step": 11256
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4787115883006293e-06,
      "loss": 1.0236,
      "step": 11257
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4772306553128477e-06,
      "loss": 1.0528,
      "step": 11258
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.475749722325065e-06,
      "loss": 1.0146,
      "step": 11259
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4742687893372827e-06,
      "loss": 1.2195,
      "step": 11260
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4727878563495006e-06,
      "loss": 1.3214,
      "step": 11261
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.471306923361718e-06,
      "loss": 1.6929,
      "step": 11262
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4698259903739357e-06,
      "loss": 1.2741,
      "step": 11263
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4683450573861536e-06,
      "loss": 1.0039,
      "step": 11264
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4668641243983715e-06,
      "loss": 1.4635,
      "step": 11265
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.465383191410589e-06,
      "loss": 1.1806,
      "step": 11266
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4639022584228066e-06,
      "loss": 1.0881,
      "step": 11267
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.462421325435024e-06,
      "loss": 1.3803,
      "step": 11268
    },
    {
      "epoch": 2.48,
      "learning_rate": 3.4609403924472416e-06,
      "loss": 1.3463,
      "step": 11269
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.45945945945946e-06,
      "loss": 1.2314,
      "step": 11270
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4579785264716775e-06,
      "loss": 1.1935,
      "step": 11271
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.456497593483895e-06,
      "loss": 1.2689,
      "step": 11272
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4550166604961125e-06,
      "loss": 1.3558,
      "step": 11273
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4535357275083305e-06,
      "loss": 1.3572,
      "step": 11274
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4520547945205484e-06,
      "loss": 0.9481,
      "step": 11275
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.450573861532766e-06,
      "loss": 1.1635,
      "step": 11276
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.449092928544984e-06,
      "loss": 1.3258,
      "step": 11277
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4476119955572014e-06,
      "loss": 1.4459,
      "step": 11278
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.446131062569419e-06,
      "loss": 1.1238,
      "step": 11279
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4446501295816364e-06,
      "loss": 1.1357,
      "step": 11280
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4431691965938548e-06,
      "loss": 1.2578,
      "step": 11281
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4416882636060723e-06,
      "loss": 1.0569,
      "step": 11282
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.44020733061829e-06,
      "loss": 1.2486,
      "step": 11283
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4387263976305073e-06,
      "loss": 1.0986,
      "step": 11284
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.437245464642725e-06,
      "loss": 1.2375,
      "step": 11285
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4357645316549432e-06,
      "loss": 1.1673,
      "step": 11286
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4342835986671607e-06,
      "loss": 1.3956,
      "step": 11287
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4328026656793782e-06,
      "loss": 1.3207,
      "step": 11288
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4313217326915958e-06,
      "loss": 1.3742,
      "step": 11289
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4298407997038137e-06,
      "loss": 1.0999,
      "step": 11290
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4283598667160312e-06,
      "loss": 1.3167,
      "step": 11291
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.426878933728249e-06,
      "loss": 1.2082,
      "step": 11292
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.425398000740467e-06,
      "loss": 1.1594,
      "step": 11293
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4239170677526846e-06,
      "loss": 0.9508,
      "step": 11294
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.422436134764902e-06,
      "loss": 1.2263,
      "step": 11295
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4209552017771197e-06,
      "loss": 1.1661,
      "step": 11296
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.419474268789337e-06,
      "loss": 1.0841,
      "step": 11297
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4179933358015555e-06,
      "loss": 1.0881,
      "step": 11298
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.416512402813773e-06,
      "loss": 1.1358,
      "step": 11299
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4150314698259906e-06,
      "loss": 1.1646,
      "step": 11300
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.413550536838208e-06,
      "loss": 1.2404,
      "step": 11301
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.412069603850426e-06,
      "loss": 1.572,
      "step": 11302
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.410588670862644e-06,
      "loss": 1.1959,
      "step": 11303
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4091077378748615e-06,
      "loss": 1.2211,
      "step": 11304
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.407626804887079e-06,
      "loss": 1.226,
      "step": 11305
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.406145871899297e-06,
      "loss": 1.3329,
      "step": 11306
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4046649389115145e-06,
      "loss": 1.176,
      "step": 11307
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.403184005923732e-06,
      "loss": 1.1513,
      "step": 11308
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.4017030729359503e-06,
      "loss": 1.4158,
      "step": 11309
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.400222139948168e-06,
      "loss": 1.232,
      "step": 11310
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.3987412069603854e-06,
      "loss": 1.0567,
      "step": 11311
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.397260273972603e-06,
      "loss": 1.269,
      "step": 11312
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.3957793409848204e-06,
      "loss": 1.193,
      "step": 11313
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.3942984079970383e-06,
      "loss": 0.9981,
      "step": 11314
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3928174750092563e-06,
      "loss": 1.4616,
      "step": 11315
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.391336542021474e-06,
      "loss": 1.3571,
      "step": 11316
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3898556090336913e-06,
      "loss": 1.1745,
      "step": 11317
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3883746760459093e-06,
      "loss": 1.2705,
      "step": 11318
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3868937430581268e-06,
      "loss": 1.1051,
      "step": 11319
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3854128100703447e-06,
      "loss": 1.2255,
      "step": 11320
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3839318770825622e-06,
      "loss": 1.2348,
      "step": 11321
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.38245094409478e-06,
      "loss": 1.2262,
      "step": 11322
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3809700111069977e-06,
      "loss": 1.1436,
      "step": 11323
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.379489078119215e-06,
      "loss": 1.3025,
      "step": 11324
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3780081451314327e-06,
      "loss": 1.1881,
      "step": 11325
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.376527212143651e-06,
      "loss": 1.0471,
      "step": 11326
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3750462791558686e-06,
      "loss": 1.469,
      "step": 11327
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.373565346168086e-06,
      "loss": 1.2651,
      "step": 11328
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3720844131803036e-06,
      "loss": 1.1649,
      "step": 11329
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3706034801925216e-06,
      "loss": 1.1639,
      "step": 11330
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3691225472047395e-06,
      "loss": 1.2688,
      "step": 11331
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.367641614216957e-06,
      "loss": 1.245,
      "step": 11332
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3661606812291746e-06,
      "loss": 1.1854,
      "step": 11333
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3646797482413925e-06,
      "loss": 1.2275,
      "step": 11334
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.36319881525361e-06,
      "loss": 1.4042,
      "step": 11335
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3617178822658275e-06,
      "loss": 1.0901,
      "step": 11336
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3602369492780455e-06,
      "loss": 1.2147,
      "step": 11337
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3587560162902634e-06,
      "loss": 1.2056,
      "step": 11338
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.357275083302481e-06,
      "loss": 1.0268,
      "step": 11339
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3557941503146984e-06,
      "loss": 1.1488,
      "step": 11340
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.354313217326916e-06,
      "loss": 1.3693,
      "step": 11341
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3528322843391335e-06,
      "loss": 1.2226,
      "step": 11342
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.351351351351352e-06,
      "loss": 1.145,
      "step": 11343
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3498704183635694e-06,
      "loss": 1.1599,
      "step": 11344
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.348389485375787e-06,
      "loss": 1.1695,
      "step": 11345
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.346908552388005e-06,
      "loss": 1.1903,
      "step": 11346
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3454276194002223e-06,
      "loss": 1.0372,
      "step": 11347
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3439466864124403e-06,
      "loss": 0.9799,
      "step": 11348
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.342465753424658e-06,
      "loss": 1.3377,
      "step": 11349
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3409848204368757e-06,
      "loss": 1.0768,
      "step": 11350
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3395038874490932e-06,
      "loss": 1.0561,
      "step": 11351
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3380229544613108e-06,
      "loss": 1.1729,
      "step": 11352
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3365420214735283e-06,
      "loss": 1.1827,
      "step": 11353
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3350610884857466e-06,
      "loss": 1.0255,
      "step": 11354
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.333580155497964e-06,
      "loss": 1.491,
      "step": 11355
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3320992225101817e-06,
      "loss": 1.3381,
      "step": 11356
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.330618289522399e-06,
      "loss": 1.3071,
      "step": 11357
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3291373565346167e-06,
      "loss": 1.3472,
      "step": 11358
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.327656423546835e-06,
      "loss": 1.5295,
      "step": 11359
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3261754905590526e-06,
      "loss": 1.0504,
      "step": 11360
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.32469455757127e-06,
      "loss": 1.093,
      "step": 11361
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.323213624583488e-06,
      "loss": 1.2908,
      "step": 11362
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.3217326915957056e-06,
      "loss": 1.2638,
      "step": 11363
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.320251758607923e-06,
      "loss": 1.3531,
      "step": 11364
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.318770825620141e-06,
      "loss": 1.1219,
      "step": 11365
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.317289892632359e-06,
      "loss": 1.4288,
      "step": 11366
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.3158089596445765e-06,
      "loss": 1.1904,
      "step": 11367
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.314328026656794e-06,
      "loss": 1.2416,
      "step": 11368
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.3128470936690115e-06,
      "loss": 1.197,
      "step": 11369
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.311366160681229e-06,
      "loss": 1.3507,
      "step": 11370
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.3098852276934474e-06,
      "loss": 1.4953,
      "step": 11371
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.308404294705665e-06,
      "loss": 1.4251,
      "step": 11372
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.3069233617178824e-06,
      "loss": 1.2033,
      "step": 11373
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.3054424287301e-06,
      "loss": 1.2654,
      "step": 11374
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.303961495742318e-06,
      "loss": 1.3499,
      "step": 11375
    },
    {
      "epoch": 2.51,
      "eval_loss": NaN,
      "eval_runtime": 4.3804,
      "eval_samples_per_second": 352.477,
      "eval_steps_per_second": 44.06,
      "step": 11375
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.302480562754536e-06,
      "loss": 1.3589,
      "step": 11376
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.3009996297667533e-06,
      "loss": 1.2355,
      "step": 11377
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2995186967789713e-06,
      "loss": 1.2658,
      "step": 11378
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.298037763791189e-06,
      "loss": 1.2996,
      "step": 11379
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2965568308034063e-06,
      "loss": 1.3257,
      "step": 11380
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.295075897815624e-06,
      "loss": 1.1789,
      "step": 11381
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.293594964827842e-06,
      "loss": 1.2682,
      "step": 11382
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2921140318400597e-06,
      "loss": 1.1301,
      "step": 11383
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2906330988522772e-06,
      "loss": 1.0211,
      "step": 11384
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2891521658644948e-06,
      "loss": 1.2153,
      "step": 11385
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2876712328767123e-06,
      "loss": 1.1714,
      "step": 11386
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2861902998889306e-06,
      "loss": 1.5368,
      "step": 11387
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.284709366901148e-06,
      "loss": 1.2684,
      "step": 11388
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2832284339133657e-06,
      "loss": 1.3317,
      "step": 11389
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.281747500925583e-06,
      "loss": 1.2294,
      "step": 11390
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.280266567937801e-06,
      "loss": 1.1625,
      "step": 11391
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2787856349500186e-06,
      "loss": 1.3484,
      "step": 11392
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2773047019622366e-06,
      "loss": 1.2623,
      "step": 11393
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2758237689744545e-06,
      "loss": 1.3628,
      "step": 11394
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.274342835986672e-06,
      "loss": 1.1316,
      "step": 11395
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2728619029988896e-06,
      "loss": 1.2975,
      "step": 11396
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.271380970011107e-06,
      "loss": 1.1484,
      "step": 11397
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2699000370233246e-06,
      "loss": 1.2694,
      "step": 11398
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.268419104035543e-06,
      "loss": 1.0939,
      "step": 11399
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2669381710477605e-06,
      "loss": 1.0893,
      "step": 11400
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.265457238059978e-06,
      "loss": 1.4017,
      "step": 11401
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2639763050721955e-06,
      "loss": 1.3075,
      "step": 11402
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2624953720844134e-06,
      "loss": 1.3101,
      "step": 11403
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.2610144390966314e-06,
      "loss": 1.2459,
      "step": 11404
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.259533506108849e-06,
      "loss": 1.1177,
      "step": 11405
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2580525731210664e-06,
      "loss": 1.0679,
      "step": 11406
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2565716401332844e-06,
      "loss": 1.1509,
      "step": 11407
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.255090707145502e-06,
      "loss": 1.215,
      "step": 11408
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2536097741577194e-06,
      "loss": 1.3105,
      "step": 11409
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2521288411699378e-06,
      "loss": 1.0441,
      "step": 11410
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2506479081821553e-06,
      "loss": 1.3836,
      "step": 11411
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.249166975194373e-06,
      "loss": 1.1068,
      "step": 11412
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2476860422065903e-06,
      "loss": 1.4035,
      "step": 11413
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.246205109218808e-06,
      "loss": 0.9171,
      "step": 11414
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2447241762310253e-06,
      "loss": 1.1968,
      "step": 11415
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2432432432432437e-06,
      "loss": 1.4463,
      "step": 11416
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2417623102554612e-06,
      "loss": 0.9947,
      "step": 11417
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2402813772676787e-06,
      "loss": 1.2328,
      "step": 11418
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2388004442798967e-06,
      "loss": 1.0372,
      "step": 11419
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.237319511292114e-06,
      "loss": 1.2412,
      "step": 11420
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.235838578304332e-06,
      "loss": 1.3182,
      "step": 11421
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2343576453165497e-06,
      "loss": 1.2243,
      "step": 11422
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2328767123287676e-06,
      "loss": 1.0506,
      "step": 11423
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.231395779340985e-06,
      "loss": 1.1702,
      "step": 11424
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2299148463532026e-06,
      "loss": 1.1636,
      "step": 11425
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.22843391336542e-06,
      "loss": 1.2231,
      "step": 11426
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2269529803776385e-06,
      "loss": 1.1031,
      "step": 11427
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.225472047389856e-06,
      "loss": 1.2933,
      "step": 11428
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2239911144020735e-06,
      "loss": 1.2215,
      "step": 11429
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.222510181414291e-06,
      "loss": 1.416,
      "step": 11430
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2210292484265086e-06,
      "loss": 1.0267,
      "step": 11431
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.219548315438727e-06,
      "loss": 1.0474,
      "step": 11432
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2180673824509445e-06,
      "loss": 1.3207,
      "step": 11433
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.216586449463162e-06,
      "loss": 1.1371,
      "step": 11434
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.21510551647538e-06,
      "loss": 1.2133,
      "step": 11435
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2136245834875974e-06,
      "loss": 1.1198,
      "step": 11436
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.212143650499815e-06,
      "loss": 1.2975,
      "step": 11437
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.210662717512033e-06,
      "loss": 1.2196,
      "step": 11438
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.209181784524251e-06,
      "loss": 1.0641,
      "step": 11439
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2077008515364683e-06,
      "loss": 1.3275,
      "step": 11440
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.206219918548686e-06,
      "loss": 1.152,
      "step": 11441
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2047389855609034e-06,
      "loss": 1.2299,
      "step": 11442
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.203258052573121e-06,
      "loss": 1.0074,
      "step": 11443
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2017771195853393e-06,
      "loss": 1.4994,
      "step": 11444
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.2002961865975568e-06,
      "loss": 1.5283,
      "step": 11445
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.1988152536097743e-06,
      "loss": 1.0596,
      "step": 11446
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.197334320621992e-06,
      "loss": 1.4796,
      "step": 11447
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.1958533876342097e-06,
      "loss": 1.2311,
      "step": 11448
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.1943724546464277e-06,
      "loss": 1.1984,
      "step": 11449
    },
    {
      "epoch": 2.52,
      "learning_rate": 3.192891521658645e-06,
      "loss": 1.1097,
      "step": 11450
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.191410588670863e-06,
      "loss": 1.1602,
      "step": 11451
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1899296556830807e-06,
      "loss": 0.9808,
      "step": 11452
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.188448722695298e-06,
      "loss": 1.2896,
      "step": 11453
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1869677897075157e-06,
      "loss": 1.3692,
      "step": 11454
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.185486856719734e-06,
      "loss": 1.1005,
      "step": 11455
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1840059237319516e-06,
      "loss": 1.0877,
      "step": 11456
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.182524990744169e-06,
      "loss": 1.3245,
      "step": 11457
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1810440577563866e-06,
      "loss": 0.952,
      "step": 11458
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.179563124768604e-06,
      "loss": 1.3577,
      "step": 11459
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1780821917808225e-06,
      "loss": 1.4219,
      "step": 11460
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.17660125879304e-06,
      "loss": 1.0214,
      "step": 11461
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1751203258052575e-06,
      "loss": 0.9542,
      "step": 11462
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.173639392817475e-06,
      "loss": 1.3378,
      "step": 11463
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.172158459829693e-06,
      "loss": 1.2232,
      "step": 11464
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1706775268419105e-06,
      "loss": 1.1798,
      "step": 11465
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1691965938541284e-06,
      "loss": 1.2226,
      "step": 11466
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1677156608663464e-06,
      "loss": 1.4027,
      "step": 11467
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.166234727878564e-06,
      "loss": 1.2364,
      "step": 11468
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1647537948907814e-06,
      "loss": 1.2708,
      "step": 11469
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.163272861902999e-06,
      "loss": 1.301,
      "step": 11470
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1617919289152164e-06,
      "loss": 1.2314,
      "step": 11471
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.160310995927435e-06,
      "loss": 1.2664,
      "step": 11472
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1588300629396523e-06,
      "loss": 1.0868,
      "step": 11473
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.15734912995187e-06,
      "loss": 1.3227,
      "step": 11474
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1558681969640874e-06,
      "loss": 1.1601,
      "step": 11475
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1543872639763053e-06,
      "loss": 0.9102,
      "step": 11476
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1529063309885232e-06,
      "loss": 0.9805,
      "step": 11477
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1514253980007408e-06,
      "loss": 1.218,
      "step": 11478
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1499444650129583e-06,
      "loss": 1.1528,
      "step": 11479
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1484635320251762e-06,
      "loss": 1.329,
      "step": 11480
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1469825990373937e-06,
      "loss": 1.3615,
      "step": 11481
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1455016660496113e-06,
      "loss": 1.3093,
      "step": 11482
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1440207330618296e-06,
      "loss": 1.2743,
      "step": 11483
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.142539800074047e-06,
      "loss": 1.3702,
      "step": 11484
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1410588670862647e-06,
      "loss": 1.3648,
      "step": 11485
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.139577934098482e-06,
      "loss": 1.3101,
      "step": 11486
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1380970011106997e-06,
      "loss": 1.0305,
      "step": 11487
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.136616068122918e-06,
      "loss": 1.3286,
      "step": 11488
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1351351351351356e-06,
      "loss": 1.2636,
      "step": 11489
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.133654202147353e-06,
      "loss": 1.2029,
      "step": 11490
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1321732691595706e-06,
      "loss": 1.3631,
      "step": 11491
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1306923361717885e-06,
      "loss": 1.0422,
      "step": 11492
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.129211403184006e-06,
      "loss": 1.3212,
      "step": 11493
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.127730470196224e-06,
      "loss": 1.4245,
      "step": 11494
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1262495372084415e-06,
      "loss": 1.297,
      "step": 11495
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.1247686042206595e-06,
      "loss": 1.2036,
      "step": 11496
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.123287671232877e-06,
      "loss": 1.124,
      "step": 11497
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.1218067382450945e-06,
      "loss": 1.2639,
      "step": 11498
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.120325805257312e-06,
      "loss": 0.9034,
      "step": 11499
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.1188448722695304e-06,
      "loss": 1.3384,
      "step": 11500
    },
    {
      "epoch": 2.54,
      "eval_loss": NaN,
      "eval_runtime": 4.3771,
      "eval_samples_per_second": 352.742,
      "eval_steps_per_second": 44.093,
      "step": 11500
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.117363939281748e-06,
      "loss": 1.1973,
      "step": 11501
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.1158830062939654e-06,
      "loss": 1.2439,
      "step": 11502
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.114402073306183e-06,
      "loss": 1.2829,
      "step": 11503
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.1129211403184004e-06,
      "loss": 1.0572,
      "step": 11504
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.111440207330619e-06,
      "loss": 1.2422,
      "step": 11505
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.1099592743428363e-06,
      "loss": 1.2629,
      "step": 11506
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.108478341355054e-06,
      "loss": 1.1353,
      "step": 11507
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.1069974083672718e-06,
      "loss": 1.1532,
      "step": 11508
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.1055164753794893e-06,
      "loss": 1.2961,
      "step": 11509
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.104035542391707e-06,
      "loss": 1.1978,
      "step": 11510
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.1025546094039247e-06,
      "loss": 1.4024,
      "step": 11511
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.1010736764161427e-06,
      "loss": 1.1179,
      "step": 11512
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.09959274342836e-06,
      "loss": 1.4211,
      "step": 11513
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0981118104405777e-06,
      "loss": 1.1822,
      "step": 11514
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0966308774527952e-06,
      "loss": 1.1204,
      "step": 11515
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0951499444650128e-06,
      "loss": 1.3391,
      "step": 11516
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.093669011477231e-06,
      "loss": 1.0337,
      "step": 11517
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0921880784894486e-06,
      "loss": 1.3358,
      "step": 11518
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.090707145501666e-06,
      "loss": 1.1179,
      "step": 11519
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0892262125138837e-06,
      "loss": 1.3743,
      "step": 11520
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0877452795261016e-06,
      "loss": 1.1996,
      "step": 11521
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0862643465383196e-06,
      "loss": 1.1723,
      "step": 11522
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.084783413550537e-06,
      "loss": 1.2592,
      "step": 11523
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.083302480562755e-06,
      "loss": 1.1784,
      "step": 11524
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0818215475749725e-06,
      "loss": 1.213,
      "step": 11525
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.08034061458719e-06,
      "loss": 1.2975,
      "step": 11526
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0788596815994076e-06,
      "loss": 1.2284,
      "step": 11527
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.077378748611626e-06,
      "loss": 1.4218,
      "step": 11528
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0758978156238434e-06,
      "loss": 1.172,
      "step": 11529
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.074416882636061e-06,
      "loss": 1.1667,
      "step": 11530
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0729359496482785e-06,
      "loss": 1.4342,
      "step": 11531
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.071455016660496e-06,
      "loss": 1.3945,
      "step": 11532
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0699740836727144e-06,
      "loss": 1.1705,
      "step": 11533
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.068493150684932e-06,
      "loss": 1.1004,
      "step": 11534
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0670122176971494e-06,
      "loss": 1.2878,
      "step": 11535
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.065531284709367e-06,
      "loss": 1.0816,
      "step": 11536
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.064050351721585e-06,
      "loss": 1.2964,
      "step": 11537
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0625694187338024e-06,
      "loss": 1.0226,
      "step": 11538
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0610884857460203e-06,
      "loss": 1.459,
      "step": 11539
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0596075527582382e-06,
      "loss": 1.1877,
      "step": 11540
    },
    {
      "epoch": 2.54,
      "learning_rate": 3.0581266197704558e-06,
      "loss": 1.1805,
      "step": 11541
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0566456867826733e-06,
      "loss": 1.1439,
      "step": 11542
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.055164753794891e-06,
      "loss": 1.192,
      "step": 11543
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0536838208071083e-06,
      "loss": 1.1235,
      "step": 11544
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0522028878193267e-06,
      "loss": 1.2781,
      "step": 11545
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.050721954831544e-06,
      "loss": 1.1937,
      "step": 11546
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0492410218437617e-06,
      "loss": 0.9338,
      "step": 11547
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0477600888559792e-06,
      "loss": 1.2309,
      "step": 11548
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.046279155868197e-06,
      "loss": 1.2193,
      "step": 11549
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.044798222880415e-06,
      "loss": 1.2154,
      "step": 11550
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0433172898926326e-06,
      "loss": 1.2504,
      "step": 11551
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.04183635690485e-06,
      "loss": 0.9864,
      "step": 11552
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.040355423917068e-06,
      "loss": 1.3453,
      "step": 11553
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0388744909292856e-06,
      "loss": 1.0505,
      "step": 11554
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.037393557941503e-06,
      "loss": 1.132,
      "step": 11555
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0359126249537215e-06,
      "loss": 1.1837,
      "step": 11556
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.034431691965939e-06,
      "loss": 1.3159,
      "step": 11557
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0329507589781565e-06,
      "loss": 1.1204,
      "step": 11558
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.031469825990374e-06,
      "loss": 1.1988,
      "step": 11559
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0299888930025915e-06,
      "loss": 1.1028,
      "step": 11560
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.02850796001481e-06,
      "loss": 1.2377,
      "step": 11561
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0270270270270274e-06,
      "loss": 1.0543,
      "step": 11562
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.025546094039245e-06,
      "loss": 1.3852,
      "step": 11563
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0240651610514625e-06,
      "loss": 1.1362,
      "step": 11564
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0225842280636804e-06,
      "loss": 1.2269,
      "step": 11565
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.021103295075898e-06,
      "loss": 1.296,
      "step": 11566
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.019622362088116e-06,
      "loss": 1.348,
      "step": 11567
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0181414291003334e-06,
      "loss": 0.946,
      "step": 11568
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0166604961125513e-06,
      "loss": 1.1575,
      "step": 11569
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.015179563124769e-06,
      "loss": 1.1698,
      "step": 11570
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0136986301369864e-06,
      "loss": 1.0751,
      "step": 11571
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.012217697149204e-06,
      "loss": 1.2361,
      "step": 11572
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0107367641614222e-06,
      "loss": 1.2104,
      "step": 11573
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0092558311736397e-06,
      "loss": 1.1679,
      "step": 11574
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0077748981858573e-06,
      "loss": 1.2297,
      "step": 11575
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0062939651980748e-06,
      "loss": 1.13,
      "step": 11576
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0048130322102927e-06,
      "loss": 1.3085,
      "step": 11577
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0033320992225107e-06,
      "loss": 1.1366,
      "step": 11578
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.001851166234728e-06,
      "loss": 1.2943,
      "step": 11579
    },
    {
      "epoch": 2.55,
      "learning_rate": 3.0003702332469457e-06,
      "loss": 1.2894,
      "step": 11580
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.9988893002591636e-06,
      "loss": 1.1636,
      "step": 11581
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.997408367271381e-06,
      "loss": 1.1352,
      "step": 11582
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.9959274342835987e-06,
      "loss": 1.238,
      "step": 11583
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.9944465012958166e-06,
      "loss": 1.2566,
      "step": 11584
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.9929655683080346e-06,
      "loss": 1.0558,
      "step": 11585
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.991484635320252e-06,
      "loss": 1.274,
      "step": 11586
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9900037023324696e-06,
      "loss": 1.1877,
      "step": 11587
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.988522769344687e-06,
      "loss": 1.067,
      "step": 11588
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9870418363569055e-06,
      "loss": 1.0708,
      "step": 11589
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.985560903369123e-06,
      "loss": 1.1747,
      "step": 11590
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9840799703813405e-06,
      "loss": 1.2715,
      "step": 11591
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.982599037393558e-06,
      "loss": 1.3356,
      "step": 11592
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.981118104405776e-06,
      "loss": 1.3547,
      "step": 11593
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9796371714179935e-06,
      "loss": 1.2512,
      "step": 11594
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9781562384302114e-06,
      "loss": 1.123,
      "step": 11595
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.976675305442429e-06,
      "loss": 1.1055,
      "step": 11596
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.975194372454647e-06,
      "loss": 1.0516,
      "step": 11597
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9737134394668644e-06,
      "loss": 1.1224,
      "step": 11598
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.972232506479082e-06,
      "loss": 1.3095,
      "step": 11599
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9707515734912994e-06,
      "loss": 1.2135,
      "step": 11600
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9692706405035178e-06,
      "loss": 1.1781,
      "step": 11601
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9677897075157353e-06,
      "loss": 1.2103,
      "step": 11602
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.966308774527953e-06,
      "loss": 1.2057,
      "step": 11603
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9648278415401703e-06,
      "loss": 1.213,
      "step": 11604
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.963346908552388e-06,
      "loss": 1.1661,
      "step": 11605
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9618659755646062e-06,
      "loss": 1.2677,
      "step": 11606
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9603850425768237e-06,
      "loss": 1.0379,
      "step": 11607
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9589041095890413e-06,
      "loss": 1.2934,
      "step": 11608
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.957423176601259e-06,
      "loss": 1.2814,
      "step": 11609
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9559422436134767e-06,
      "loss": 1.1733,
      "step": 11610
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9544613106256942e-06,
      "loss": 1.2177,
      "step": 11611
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.952980377637912e-06,
      "loss": 1.2374,
      "step": 11612
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.95149944465013e-06,
      "loss": 1.4183,
      "step": 11613
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9500185116623476e-06,
      "loss": 1.2007,
      "step": 11614
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.948537578674565e-06,
      "loss": 1.1264,
      "step": 11615
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9470566456867827e-06,
      "loss": 1.1797,
      "step": 11616
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.945575712699e-06,
      "loss": 1.343,
      "step": 11617
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9440947797112185e-06,
      "loss": 1.0237,
      "step": 11618
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.942613846723436e-06,
      "loss": 1.2406,
      "step": 11619
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9411329137356536e-06,
      "loss": 1.1662,
      "step": 11620
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.939651980747871e-06,
      "loss": 1.1835,
      "step": 11621
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.938171047760089e-06,
      "loss": 1.0309,
      "step": 11622
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.936690114772307e-06,
      "loss": 1.2321,
      "step": 11623
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9352091817845245e-06,
      "loss": 1.4667,
      "step": 11624
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9337282487967424e-06,
      "loss": 1.1232,
      "step": 11625
    },
    {
      "epoch": 2.56,
      "eval_loss": NaN,
      "eval_runtime": 4.3692,
      "eval_samples_per_second": 353.38,
      "eval_steps_per_second": 44.173,
      "step": 11625
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.93224731580896e-06,
      "loss": 1.3116,
      "step": 11626
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9307663828211775e-06,
      "loss": 1.1329,
      "step": 11627
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.929285449833395e-06,
      "loss": 1.106,
      "step": 11628
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9278045168456133e-06,
      "loss": 1.2707,
      "step": 11629
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.926323583857831e-06,
      "loss": 1.2906,
      "step": 11630
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.9248426508700484e-06,
      "loss": 1.2978,
      "step": 11631
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.923361717882266e-06,
      "loss": 1.2716,
      "step": 11632
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9218807848944834e-06,
      "loss": 1.3617,
      "step": 11633
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9203998519067018e-06,
      "loss": 1.0883,
      "step": 11634
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9189189189189193e-06,
      "loss": 1.1677,
      "step": 11635
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.917437985931137e-06,
      "loss": 1.2326,
      "step": 11636
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9159570529433543e-06,
      "loss": 1.0922,
      "step": 11637
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9144761199555723e-06,
      "loss": 0.9987,
      "step": 11638
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9129951869677898e-06,
      "loss": 1.1856,
      "step": 11639
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9115142539800077e-06,
      "loss": 1.3633,
      "step": 11640
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9100333209922257e-06,
      "loss": 1.0952,
      "step": 11641
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.908552388004443e-06,
      "loss": 1.2583,
      "step": 11642
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9070714550166607e-06,
      "loss": 1.061,
      "step": 11643
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9055905220288782e-06,
      "loss": 1.0991,
      "step": 11644
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9041095890410957e-06,
      "loss": 1.0777,
      "step": 11645
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.902628656053314e-06,
      "loss": 1.2198,
      "step": 11646
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.9011477230655316e-06,
      "loss": 1.2163,
      "step": 11647
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.899666790077749e-06,
      "loss": 1.1962,
      "step": 11648
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8981858570899666e-06,
      "loss": 1.4098,
      "step": 11649
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8967049241021846e-06,
      "loss": 1.0937,
      "step": 11650
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8952239911144025e-06,
      "loss": 1.3536,
      "step": 11651
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.89374305812662e-06,
      "loss": 1.0286,
      "step": 11652
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8922621251388376e-06,
      "loss": 1.1441,
      "step": 11653
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8907811921510555e-06,
      "loss": 1.1789,
      "step": 11654
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.889300259163273e-06,
      "loss": 1.2527,
      "step": 11655
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8878193261754905e-06,
      "loss": 1.3035,
      "step": 11656
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.886338393187709e-06,
      "loss": 0.9939,
      "step": 11657
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8848574601999264e-06,
      "loss": 1.1123,
      "step": 11658
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.883376527212144e-06,
      "loss": 1.1234,
      "step": 11659
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8818955942243614e-06,
      "loss": 1.3594,
      "step": 11660
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.880414661236579e-06,
      "loss": 1.1078,
      "step": 11661
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8789337282487973e-06,
      "loss": 1.1252,
      "step": 11662
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.877452795261015e-06,
      "loss": 1.2196,
      "step": 11663
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8759718622732324e-06,
      "loss": 1.1684,
      "step": 11664
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.87449092928545e-06,
      "loss": 1.0803,
      "step": 11665
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.873009996297668e-06,
      "loss": 1.1373,
      "step": 11666
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8715290633098853e-06,
      "loss": 1.3142,
      "step": 11667
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8700481303221033e-06,
      "loss": 1.0749,
      "step": 11668
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.868567197334321e-06,
      "loss": 1.1564,
      "step": 11669
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8670862643465387e-06,
      "loss": 1.1482,
      "step": 11670
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8656053313587563e-06,
      "loss": 1.1866,
      "step": 11671
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8641243983709738e-06,
      "loss": 1.2119,
      "step": 11672
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8626434653831913e-06,
      "loss": 1.1499,
      "step": 11673
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8611625323954096e-06,
      "loss": 1.083,
      "step": 11674
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.859681599407627e-06,
      "loss": 1.2117,
      "step": 11675
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8582006664198447e-06,
      "loss": 1.3582,
      "step": 11676
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.856719733432062e-06,
      "loss": 1.2833,
      "step": 11677
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8552388004442797e-06,
      "loss": 1.1571,
      "step": 11678
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.853757867456498e-06,
      "loss": 1.2512,
      "step": 11679
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8522769344687156e-06,
      "loss": 1.0664,
      "step": 11680
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.850796001480933e-06,
      "loss": 1.3393,
      "step": 11681
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.849315068493151e-06,
      "loss": 0.9754,
      "step": 11682
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8478341355053686e-06,
      "loss": 1.2405,
      "step": 11683
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.846353202517586e-06,
      "loss": 1.0475,
      "step": 11684
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.844872269529804e-06,
      "loss": 1.0231,
      "step": 11685
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.843391336542022e-06,
      "loss": 1.2448,
      "step": 11686
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8419104035542395e-06,
      "loss": 1.2038,
      "step": 11687
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.840429470566457e-06,
      "loss": 1.2294,
      "step": 11688
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8389485375786745e-06,
      "loss": 1.1583,
      "step": 11689
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.837467604590893e-06,
      "loss": 1.2914,
      "step": 11690
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8359866716031104e-06,
      "loss": 1.2479,
      "step": 11691
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.834505738615328e-06,
      "loss": 0.9567,
      "step": 11692
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8330248056275454e-06,
      "loss": 1.2038,
      "step": 11693
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.831543872639763e-06,
      "loss": 1.0883,
      "step": 11694
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.830062939651981e-06,
      "loss": 1.2308,
      "step": 11695
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.828582006664199e-06,
      "loss": 1.212,
      "step": 11696
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8271010736764163e-06,
      "loss": 1.1578,
      "step": 11697
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8256201406886343e-06,
      "loss": 1.3749,
      "step": 11698
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.824139207700852e-06,
      "loss": 1.3537,
      "step": 11699
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8226582747130693e-06,
      "loss": 1.2855,
      "step": 11700
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.821177341725287e-06,
      "loss": 1.3078,
      "step": 11701
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.819696408737505e-06,
      "loss": 1.2664,
      "step": 11702
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8182154757497227e-06,
      "loss": 1.1394,
      "step": 11703
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8167345427619402e-06,
      "loss": 1.0844,
      "step": 11704
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8152536097741578e-06,
      "loss": 1.2455,
      "step": 11705
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8137726767863753e-06,
      "loss": 1.14,
      "step": 11706
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8122917437985936e-06,
      "loss": 1.2077,
      "step": 11707
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.810810810810811e-06,
      "loss": 1.1157,
      "step": 11708
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8093298778230287e-06,
      "loss": 1.2482,
      "step": 11709
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.807848944835246e-06,
      "loss": 1.2715,
      "step": 11710
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.806368011847464e-06,
      "loss": 1.2459,
      "step": 11711
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8048870788596816e-06,
      "loss": 1.0568,
      "step": 11712
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8034061458718996e-06,
      "loss": 1.1617,
      "step": 11713
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.8019252128841175e-06,
      "loss": 1.1718,
      "step": 11714
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.800444279896335e-06,
      "loss": 1.1489,
      "step": 11715
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.7989633469085526e-06,
      "loss": 1.1454,
      "step": 11716
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.79748241392077e-06,
      "loss": 1.4094,
      "step": 11717
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.7960014809329884e-06,
      "loss": 1.1573,
      "step": 11718
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.794520547945206e-06,
      "loss": 1.0079,
      "step": 11719
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.7930396149574235e-06,
      "loss": 1.1891,
      "step": 11720
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.791558681969641e-06,
      "loss": 1.151,
      "step": 11721
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.7900777489818585e-06,
      "loss": 1.3474,
      "step": 11722
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7885968159940764e-06,
      "loss": 1.0455,
      "step": 11723
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7871158830062944e-06,
      "loss": 1.2548,
      "step": 11724
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.785634950018512e-06,
      "loss": 1.3304,
      "step": 11725
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7841540170307294e-06,
      "loss": 1.0741,
      "step": 11726
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7826730840429474e-06,
      "loss": 1.0621,
      "step": 11727
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.781192151055165e-06,
      "loss": 1.2419,
      "step": 11728
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7797112180673824e-06,
      "loss": 1.1424,
      "step": 11729
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7782302850796008e-06,
      "loss": 0.9812,
      "step": 11730
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7767493520918183e-06,
      "loss": 1.2783,
      "step": 11731
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.775268419104036e-06,
      "loss": 1.1065,
      "step": 11732
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7737874861162533e-06,
      "loss": 1.1048,
      "step": 11733
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.772306553128471e-06,
      "loss": 1.1808,
      "step": 11734
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.770825620140689e-06,
      "loss": 1.2702,
      "step": 11735
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7693446871529067e-06,
      "loss": 1.2903,
      "step": 11736
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7678637541651242e-06,
      "loss": 1.3185,
      "step": 11737
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7663828211773417e-06,
      "loss": 1.1315,
      "step": 11738
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7649018881895597e-06,
      "loss": 1.2934,
      "step": 11739
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.763420955201777e-06,
      "loss": 1.0239,
      "step": 11740
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.761940022213995e-06,
      "loss": 0.9035,
      "step": 11741
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7604590892262127e-06,
      "loss": 1.4434,
      "step": 11742
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7589781562384306e-06,
      "loss": 1.1599,
      "step": 11743
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.757497223250648e-06,
      "loss": 1.0454,
      "step": 11744
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7560162902628656e-06,
      "loss": 1.1863,
      "step": 11745
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.754535357275083e-06,
      "loss": 1.2279,
      "step": 11746
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7530544242873015e-06,
      "loss": 1.232,
      "step": 11747
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.751573491299519e-06,
      "loss": 1.2563,
      "step": 11748
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7500925583117365e-06,
      "loss": 1.0933,
      "step": 11749
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.748611625323954e-06,
      "loss": 1.4464,
      "step": 11750
    },
    {
      "epoch": 2.59,
      "eval_loss": NaN,
      "eval_runtime": 4.3716,
      "eval_samples_per_second": 353.187,
      "eval_steps_per_second": 44.148,
      "step": 11750
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.747130692336172e-06,
      "loss": 1.0405,
      "step": 11751
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.74564975934839e-06,
      "loss": 1.1475,
      "step": 11752
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7441688263606075e-06,
      "loss": 1.3892,
      "step": 11753
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.742687893372825e-06,
      "loss": 1.1102,
      "step": 11754
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.741206960385043e-06,
      "loss": 1.0525,
      "step": 11755
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7397260273972604e-06,
      "loss": 1.3024,
      "step": 11756
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.738245094409478e-06,
      "loss": 1.31,
      "step": 11757
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.736764161421696e-06,
      "loss": 1.1878,
      "step": 11758
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.735283228433914e-06,
      "loss": 1.2963,
      "step": 11759
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7338022954461313e-06,
      "loss": 1.2364,
      "step": 11760
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.732321362458349e-06,
      "loss": 1.1111,
      "step": 11761
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7308404294705664e-06,
      "loss": 1.3947,
      "step": 11762
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7293594964827847e-06,
      "loss": 0.8927,
      "step": 11763
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7278785634950023e-06,
      "loss": 1.2912,
      "step": 11764
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7263976305072198e-06,
      "loss": 1.2317,
      "step": 11765
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7249166975194373e-06,
      "loss": 1.0377,
      "step": 11766
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7234357645316552e-06,
      "loss": 1.3035,
      "step": 11767
    },
    {
      "epoch": 2.59,
      "learning_rate": 2.7219548315438728e-06,
      "loss": 1.0845,
      "step": 11768
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.7204738985560907e-06,
      "loss": 1.2123,
      "step": 11769
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.7189929655683082e-06,
      "loss": 1.0525,
      "step": 11770
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.717512032580526e-06,
      "loss": 1.2755,
      "step": 11771
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.7160310995927437e-06,
      "loss": 1.1745,
      "step": 11772
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.714550166604961e-06,
      "loss": 1.3283,
      "step": 11773
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.7130692336171787e-06,
      "loss": 0.9182,
      "step": 11774
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.711588300629397e-06,
      "loss": 1.0901,
      "step": 11775
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.7101073676416146e-06,
      "loss": 1.2025,
      "step": 11776
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.708626434653832e-06,
      "loss": 1.1363,
      "step": 11777
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.7071455016660496e-06,
      "loss": 1.1063,
      "step": 11778
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.705664568678267e-06,
      "loss": 1.0921,
      "step": 11779
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.7041836356904855e-06,
      "loss": 1.2185,
      "step": 11780
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.702702702702703e-06,
      "loss": 1.1104,
      "step": 11781
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.7012217697149205e-06,
      "loss": 1.1553,
      "step": 11782
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6997408367271385e-06,
      "loss": 1.2681,
      "step": 11783
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.698259903739356e-06,
      "loss": 1.2642,
      "step": 11784
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6967789707515735e-06,
      "loss": 1.2376,
      "step": 11785
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6952980377637914e-06,
      "loss": 1.0814,
      "step": 11786
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6938171047760094e-06,
      "loss": 1.2969,
      "step": 11787
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.692336171788227e-06,
      "loss": 1.3636,
      "step": 11788
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6908552388004444e-06,
      "loss": 1.3031,
      "step": 11789
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.689374305812662e-06,
      "loss": 1.3394,
      "step": 11790
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6878933728248803e-06,
      "loss": 1.2182,
      "step": 11791
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.686412439837098e-06,
      "loss": 1.0639,
      "step": 11792
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6849315068493153e-06,
      "loss": 1.2041,
      "step": 11793
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.683450573861533e-06,
      "loss": 0.9766,
      "step": 11794
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6819696408737504e-06,
      "loss": 1.1301,
      "step": 11795
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6804887078859683e-06,
      "loss": 1.3558,
      "step": 11796
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6790077748981863e-06,
      "loss": 1.0363,
      "step": 11797
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6775268419104038e-06,
      "loss": 1.2658,
      "step": 11798
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6760459089226217e-06,
      "loss": 1.1056,
      "step": 11799
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6745649759348392e-06,
      "loss": 1.2464,
      "step": 11800
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6730840429470567e-06,
      "loss": 1.1752,
      "step": 11801
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6716031099592743e-06,
      "loss": 1.2344,
      "step": 11802
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6701221769714926e-06,
      "loss": 1.1435,
      "step": 11803
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.66864124398371e-06,
      "loss": 1.4092,
      "step": 11804
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6671603109959277e-06,
      "loss": 1.2058,
      "step": 11805
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.665679378008145e-06,
      "loss": 1.2218,
      "step": 11806
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6641984450203627e-06,
      "loss": 1.367,
      "step": 11807
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.662717512032581e-06,
      "loss": 1.2229,
      "step": 11808
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6612365790447986e-06,
      "loss": 1.0218,
      "step": 11809
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.659755646057016e-06,
      "loss": 1.1156,
      "step": 11810
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6582747130692336e-06,
      "loss": 1.237,
      "step": 11811
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.6567937800814515e-06,
      "loss": 1.2169,
      "step": 11812
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.655312847093669e-06,
      "loss": 1.2173,
      "step": 11813
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.653831914105887e-06,
      "loss": 1.3394,
      "step": 11814
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.652350981118105e-06,
      "loss": 1.0256,
      "step": 11815
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6508700481303225e-06,
      "loss": 1.2692,
      "step": 11816
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.64938911514254e-06,
      "loss": 1.5195,
      "step": 11817
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6479081821547575e-06,
      "loss": 1.3009,
      "step": 11818
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.646427249166976e-06,
      "loss": 1.0561,
      "step": 11819
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6449463161791934e-06,
      "loss": 1.1758,
      "step": 11820
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.643465383191411e-06,
      "loss": 1.208,
      "step": 11821
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6419844502036284e-06,
      "loss": 1.1534,
      "step": 11822
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.640503517215846e-06,
      "loss": 1.1077,
      "step": 11823
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.639022584228064e-06,
      "loss": 1.2478,
      "step": 11824
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.637541651240282e-06,
      "loss": 1.1183,
      "step": 11825
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6360607182524993e-06,
      "loss": 1.1375,
      "step": 11826
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.634579785264717e-06,
      "loss": 1.0633,
      "step": 11827
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6330988522769348e-06,
      "loss": 1.2768,
      "step": 11828
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6316179192891523e-06,
      "loss": 1.3854,
      "step": 11829
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.63013698630137e-06,
      "loss": 1.128,
      "step": 11830
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.628656053313588e-06,
      "loss": 1.158,
      "step": 11831
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6271751203258057e-06,
      "loss": 1.0503,
      "step": 11832
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.625694187338023e-06,
      "loss": 1.1729,
      "step": 11833
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6242132543502407e-06,
      "loss": 1.1852,
      "step": 11834
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6227323213624582e-06,
      "loss": 1.0399,
      "step": 11835
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6212513883746766e-06,
      "loss": 1.1942,
      "step": 11836
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.619770455386894e-06,
      "loss": 1.1248,
      "step": 11837
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6182895223991116e-06,
      "loss": 1.3751,
      "step": 11838
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.616808589411329e-06,
      "loss": 1.0262,
      "step": 11839
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.615327656423547e-06,
      "loss": 1.526,
      "step": 11840
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6138467234357646e-06,
      "loss": 1.114,
      "step": 11841
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6123657904479826e-06,
      "loss": 0.928,
      "step": 11842
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6108848574602e-06,
      "loss": 1.0584,
      "step": 11843
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.609403924472418e-06,
      "loss": 1.254,
      "step": 11844
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6079229914846355e-06,
      "loss": 1.1497,
      "step": 11845
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.606442058496853e-06,
      "loss": 1.284,
      "step": 11846
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6049611255090706e-06,
      "loss": 1.2221,
      "step": 11847
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.603480192521289e-06,
      "loss": 1.4023,
      "step": 11848
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6019992595335064e-06,
      "loss": 1.0695,
      "step": 11849
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.600518326545724e-06,
      "loss": 1.1012,
      "step": 11850
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.5990373935579415e-06,
      "loss": 1.2592,
      "step": 11851
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.597556460570159e-06,
      "loss": 1.0543,
      "step": 11852
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.5960755275823774e-06,
      "loss": 1.2247,
      "step": 11853
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.594594594594595e-06,
      "loss": 1.4724,
      "step": 11854
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.5931136616068124e-06,
      "loss": 1.2565,
      "step": 11855
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.5916327286190303e-06,
      "loss": 1.1975,
      "step": 11856
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.590151795631248e-06,
      "loss": 1.1305,
      "step": 11857
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.5886708626434654e-06,
      "loss": 1.1324,
      "step": 11858
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.5871899296556833e-06,
      "loss": 1.3412,
      "step": 11859
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5857089966679012e-06,
      "loss": 1.0788,
      "step": 11860
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5842280636801188e-06,
      "loss": 1.1275,
      "step": 11861
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5827471306923363e-06,
      "loss": 1.3293,
      "step": 11862
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.581266197704554e-06,
      "loss": 1.2228,
      "step": 11863
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.579785264716772e-06,
      "loss": 1.2726,
      "step": 11864
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5783043317289897e-06,
      "loss": 1.173,
      "step": 11865
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.576823398741207e-06,
      "loss": 1.2496,
      "step": 11866
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5753424657534247e-06,
      "loss": 1.0887,
      "step": 11867
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5738615327656422e-06,
      "loss": 1.1094,
      "step": 11868
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.57238059977786e-06,
      "loss": 1.1013,
      "step": 11869
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.570899666790078e-06,
      "loss": 1.1281,
      "step": 11870
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5694187338022956e-06,
      "loss": 1.1705,
      "step": 11871
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5679378008145136e-06,
      "loss": 1.4076,
      "step": 11872
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.566456867826731e-06,
      "loss": 1.1373,
      "step": 11873
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5649759348389486e-06,
      "loss": 1.2402,
      "step": 11874
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.563495001851166e-06,
      "loss": 1.2563,
      "step": 11875
    },
    {
      "epoch": 2.62,
      "eval_loss": NaN,
      "eval_runtime": 4.3785,
      "eval_samples_per_second": 352.632,
      "eval_steps_per_second": 44.079,
      "step": 11875
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5620140688633845e-06,
      "loss": 1.1474,
      "step": 11876
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.560533135875602e-06,
      "loss": 1.2627,
      "step": 11877
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5590522028878195e-06,
      "loss": 0.9115,
      "step": 11878
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.557571269900037e-06,
      "loss": 1.3394,
      "step": 11879
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5560903369122546e-06,
      "loss": 1.2561,
      "step": 11880
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.554609403924473e-06,
      "loss": 1.2696,
      "step": 11881
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5531284709366904e-06,
      "loss": 1.1539,
      "step": 11882
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.551647537948908e-06,
      "loss": 1.227,
      "step": 11883
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5501666049611255e-06,
      "loss": 1.2625,
      "step": 11884
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5486856719733434e-06,
      "loss": 1.4214,
      "step": 11885
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.547204738985561e-06,
      "loss": 1.2849,
      "step": 11886
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.545723805997779e-06,
      "loss": 1.0367,
      "step": 11887
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.544242873009997e-06,
      "loss": 1.1143,
      "step": 11888
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5427619400222143e-06,
      "loss": 1.2652,
      "step": 11889
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.541281007034432e-06,
      "loss": 1.0874,
      "step": 11890
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5398000740466494e-06,
      "loss": 1.2697,
      "step": 11891
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5383191410588677e-06,
      "loss": 1.2892,
      "step": 11892
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5368382080710852e-06,
      "loss": 1.0229,
      "step": 11893
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5353572750833028e-06,
      "loss": 1.174,
      "step": 11894
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5338763420955203e-06,
      "loss": 1.176,
      "step": 11895
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5323954091077378e-06,
      "loss": 1.1801,
      "step": 11896
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5309144761199557e-06,
      "loss": 1.448,
      "step": 11897
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5294335431321737e-06,
      "loss": 1.0795,
      "step": 11898
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.527952610144391e-06,
      "loss": 1.1621,
      "step": 11899
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5264716771566087e-06,
      "loss": 1.021,
      "step": 11900
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5249907441688266e-06,
      "loss": 1.2045,
      "step": 11901
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.523509811181044e-06,
      "loss": 1.3039,
      "step": 11902
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.5220288781932617e-06,
      "loss": 1.1353,
      "step": 11903
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.52054794520548e-06,
      "loss": 1.2634,
      "step": 11904
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.5190670122176976e-06,
      "loss": 1.0938,
      "step": 11905
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.517586079229915e-06,
      "loss": 1.3019,
      "step": 11906
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.5161051462421326e-06,
      "loss": 1.2687,
      "step": 11907
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.51462421325435e-06,
      "loss": 1.1241,
      "step": 11908
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.5131432802665685e-06,
      "loss": 1.1476,
      "step": 11909
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.511662347278786e-06,
      "loss": 1.3331,
      "step": 11910
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.5101814142910035e-06,
      "loss": 1.3977,
      "step": 11911
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.508700481303221e-06,
      "loss": 1.2301,
      "step": 11912
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.507219548315439e-06,
      "loss": 1.0623,
      "step": 11913
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.5057386153276565e-06,
      "loss": 1.2243,
      "step": 11914
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.5042576823398744e-06,
      "loss": 0.9389,
      "step": 11915
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.502776749352092e-06,
      "loss": 1.1984,
      "step": 11916
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.50129581636431e-06,
      "loss": 1.133,
      "step": 11917
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4998148833765274e-06,
      "loss": 1.0801,
      "step": 11918
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4983339503887453e-06,
      "loss": 1.2604,
      "step": 11919
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.496853017400963e-06,
      "loss": 1.2705,
      "step": 11920
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4953720844131804e-06,
      "loss": 1.1205,
      "step": 11921
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4938911514253983e-06,
      "loss": 1.2129,
      "step": 11922
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.492410218437616e-06,
      "loss": 1.1006,
      "step": 11923
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4909292854498333e-06,
      "loss": 1.1448,
      "step": 11924
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4894483524620513e-06,
      "loss": 1.2874,
      "step": 11925
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.487967419474269e-06,
      "loss": 1.2179,
      "step": 11926
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4864864864864867e-06,
      "loss": 1.017,
      "step": 11927
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4850055534987043e-06,
      "loss": 0.9926,
      "step": 11928
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.483524620510922e-06,
      "loss": 1.2514,
      "step": 11929
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4820436875231397e-06,
      "loss": 1.1582,
      "step": 11930
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4805627545353577e-06,
      "loss": 1.2902,
      "step": 11931
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.479081821547575e-06,
      "loss": 1.2964,
      "step": 11932
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.477600888559793e-06,
      "loss": 1.2639,
      "step": 11933
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4761199555720106e-06,
      "loss": 1.4487,
      "step": 11934
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.474639022584228e-06,
      "loss": 1.2887,
      "step": 11935
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.473158089596446e-06,
      "loss": 1.3932,
      "step": 11936
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4716771566086636e-06,
      "loss": 1.1155,
      "step": 11937
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.470196223620881e-06,
      "loss": 1.138,
      "step": 11938
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.468715290633099e-06,
      "loss": 1.1069,
      "step": 11939
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4672343576453166e-06,
      "loss": 1.2498,
      "step": 11940
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4657534246575345e-06,
      "loss": 1.0283,
      "step": 11941
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.464272491669752e-06,
      "loss": 1.1624,
      "step": 11942
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.46279155868197e-06,
      "loss": 1.2876,
      "step": 11943
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4613106256941875e-06,
      "loss": 1.2157,
      "step": 11944
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4598296927064054e-06,
      "loss": 1.1155,
      "step": 11945
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.458348759718623e-06,
      "loss": 1.1314,
      "step": 11946
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.456867826730841e-06,
      "loss": 0.9883,
      "step": 11947
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.4553868937430584e-06,
      "loss": 1.1296,
      "step": 11948
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.453905960755276e-06,
      "loss": 1.299,
      "step": 11949
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.452425027767494e-06,
      "loss": 1.3024,
      "step": 11950
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4509440947797114e-06,
      "loss": 1.24,
      "step": 11951
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.449463161791929e-06,
      "loss": 1.193,
      "step": 11952
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.447982228804147e-06,
      "loss": 1.3483,
      "step": 11953
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4465012958163644e-06,
      "loss": 1.0752,
      "step": 11954
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.445020362828582e-06,
      "loss": 1.1961,
      "step": 11955
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4435394298408e-06,
      "loss": 1.2223,
      "step": 11956
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4420584968530178e-06,
      "loss": 1.1306,
      "step": 11957
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4405775638652353e-06,
      "loss": 1.175,
      "step": 11958
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.439096630877453e-06,
      "loss": 1.4637,
      "step": 11959
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4376156978896707e-06,
      "loss": 1.246,
      "step": 11960
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4361347649018887e-06,
      "loss": 1.3156,
      "step": 11961
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.434653831914106e-06,
      "loss": 1.1407,
      "step": 11962
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4331728989263237e-06,
      "loss": 1.2063,
      "step": 11963
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4316919659385416e-06,
      "loss": 1.3369,
      "step": 11964
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.430211032950759e-06,
      "loss": 1.2264,
      "step": 11965
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4287300999629767e-06,
      "loss": 1.3107,
      "step": 11966
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4272491669751946e-06,
      "loss": 1.1466,
      "step": 11967
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.425768233987412e-06,
      "loss": 1.2362,
      "step": 11968
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4242873009996296e-06,
      "loss": 1.1518,
      "step": 11969
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4228063680118476e-06,
      "loss": 1.1616,
      "step": 11970
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.421325435024065e-06,
      "loss": 1.0715,
      "step": 11971
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.419844502036283e-06,
      "loss": 1.1342,
      "step": 11972
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.418363569048501e-06,
      "loss": 1.232,
      "step": 11973
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4168826360607185e-06,
      "loss": 1.4141,
      "step": 11974
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4154017030729364e-06,
      "loss": 1.2174,
      "step": 11975
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.413920770085154e-06,
      "loss": 1.1178,
      "step": 11976
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4124398370973715e-06,
      "loss": 1.3323,
      "step": 11977
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4109589041095894e-06,
      "loss": 1.2852,
      "step": 11978
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.409477971121807e-06,
      "loss": 0.9927,
      "step": 11979
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4079970381340245e-06,
      "loss": 1.3587,
      "step": 11980
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4065161051462424e-06,
      "loss": 1.2439,
      "step": 11981
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.40503517215846e-06,
      "loss": 1.1964,
      "step": 11982
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4035542391706774e-06,
      "loss": 1.2226,
      "step": 11983
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.4020733061828954e-06,
      "loss": 1.4816,
      "step": 11984
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.400592373195113e-06,
      "loss": 1.1553,
      "step": 11985
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.399111440207331e-06,
      "loss": 1.3049,
      "step": 11986
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.3976305072195483e-06,
      "loss": 1.3084,
      "step": 11987
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.3961495742317663e-06,
      "loss": 1.2722,
      "step": 11988
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.3946686412439842e-06,
      "loss": 1.5505,
      "step": 11989
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.3931877082562017e-06,
      "loss": 1.2415,
      "step": 11990
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.3917067752684193e-06,
      "loss": 1.0918,
      "step": 11991
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.390225842280637e-06,
      "loss": 1.2605,
      "step": 11992
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.3887449092928547e-06,
      "loss": 1.2991,
      "step": 11993
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.3872639763050722e-06,
      "loss": 1.0912,
      "step": 11994
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.38578304331729e-06,
      "loss": 1.2864,
      "step": 11995
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3843021103295077e-06,
      "loss": 1.2299,
      "step": 11996
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.382821177341725e-06,
      "loss": 1.1588,
      "step": 11997
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.381340244353943e-06,
      "loss": 1.2523,
      "step": 11998
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3798593113661607e-06,
      "loss": 1.3058,
      "step": 11999
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3783783783783786e-06,
      "loss": 1.0987,
      "step": 12000
    },
    {
      "epoch": 2.65,
      "eval_loss": NaN,
      "eval_runtime": 4.3707,
      "eval_samples_per_second": 353.264,
      "eval_steps_per_second": 44.158,
      "step": 12000
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.376897445390596e-06,
      "loss": 1.2846,
      "step": 12001
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.375416512402814e-06,
      "loss": 1.2241,
      "step": 12002
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3739355794150316e-06,
      "loss": 1.2008,
      "step": 12003
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3724546464272495e-06,
      "loss": 1.0729,
      "step": 12004
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.370973713439467e-06,
      "loss": 1.2038,
      "step": 12005
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.369492780451685e-06,
      "loss": 1.5049,
      "step": 12006
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3680118474639025e-06,
      "loss": 1.0288,
      "step": 12007
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.36653091447612e-06,
      "loss": 1.2902,
      "step": 12008
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.365049981488338e-06,
      "loss": 1.2179,
      "step": 12009
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3635690485005555e-06,
      "loss": 1.1274,
      "step": 12010
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.362088115512773e-06,
      "loss": 1.3204,
      "step": 12011
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.360607182524991e-06,
      "loss": 1.3105,
      "step": 12012
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3591262495372084e-06,
      "loss": 1.2384,
      "step": 12013
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3576453165494264e-06,
      "loss": 1.2822,
      "step": 12014
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.356164383561644e-06,
      "loss": 1.2195,
      "step": 12015
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.354683450573862e-06,
      "loss": 1.3622,
      "step": 12016
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3532025175860794e-06,
      "loss": 1.2725,
      "step": 12017
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3517215845982973e-06,
      "loss": 1.3757,
      "step": 12018
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.350240651610515e-06,
      "loss": 1.3043,
      "step": 12019
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3487597186227328e-06,
      "loss": 1.2517,
      "step": 12020
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3472787856349503e-06,
      "loss": 1.6843,
      "step": 12021
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3457978526471678e-06,
      "loss": 1.1624,
      "step": 12022
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3443169196593857e-06,
      "loss": 1.2125,
      "step": 12023
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3428359866716032e-06,
      "loss": 1.2419,
      "step": 12024
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3413550536838208e-06,
      "loss": 1.216,
      "step": 12025
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3398741206960387e-06,
      "loss": 1.0603,
      "step": 12026
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3383931877082562e-06,
      "loss": 1.282,
      "step": 12027
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.336912254720474e-06,
      "loss": 1.2091,
      "step": 12028
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3354313217326917e-06,
      "loss": 1.3214,
      "step": 12029
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3339503887449096e-06,
      "loss": 1.1347,
      "step": 12030
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.332469455757127e-06,
      "loss": 1.174,
      "step": 12031
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.330988522769345e-06,
      "loss": 1.1972,
      "step": 12032
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3295075897815626e-06,
      "loss": 1.3224,
      "step": 12033
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3280266567937805e-06,
      "loss": 1.2585,
      "step": 12034
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.326545723805998e-06,
      "loss": 1.1899,
      "step": 12035
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3250647908182156e-06,
      "loss": 1.213,
      "step": 12036
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3235838578304335e-06,
      "loss": 1.3178,
      "step": 12037
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.322102924842651e-06,
      "loss": 1.2524,
      "step": 12038
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3206219918548685e-06,
      "loss": 1.1412,
      "step": 12039
    },
    {
      "epoch": 2.65,
      "learning_rate": 2.3191410588670865e-06,
      "loss": 1.3625,
      "step": 12040
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.317660125879304e-06,
      "loss": 1.1047,
      "step": 12041
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.3161791928915215e-06,
      "loss": 1.1615,
      "step": 12042
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.3146982599037395e-06,
      "loss": 1.1702,
      "step": 12043
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.3132173269159574e-06,
      "loss": 1.3324,
      "step": 12044
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.311736393928175e-06,
      "loss": 1.1021,
      "step": 12045
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.310255460940393e-06,
      "loss": 1.0725,
      "step": 12046
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.3087745279526104e-06,
      "loss": 1.1628,
      "step": 12047
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.3072935949648283e-06,
      "loss": 0.9952,
      "step": 12048
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.305812661977046e-06,
      "loss": 1.1272,
      "step": 12049
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.3043317289892633e-06,
      "loss": 1.0596,
      "step": 12050
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.3028507960014813e-06,
      "loss": 0.9286,
      "step": 12051
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.301369863013699e-06,
      "loss": 1.2965,
      "step": 12052
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2998889300259163e-06,
      "loss": 1.2072,
      "step": 12053
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2984079970381343e-06,
      "loss": 1.0765,
      "step": 12054
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2969270640503518e-06,
      "loss": 1.3329,
      "step": 12055
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2954461310625693e-06,
      "loss": 1.2572,
      "step": 12056
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2939651980747872e-06,
      "loss": 1.526,
      "step": 12057
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2924842650870047e-06,
      "loss": 1.4841,
      "step": 12058
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2910033320992227e-06,
      "loss": 1.2822,
      "step": 12059
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2895223991114406e-06,
      "loss": 1.4393,
      "step": 12060
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.288041466123658e-06,
      "loss": 1.2442,
      "step": 12061
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.286560533135876e-06,
      "loss": 1.1748,
      "step": 12062
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2850796001480936e-06,
      "loss": 1.2937,
      "step": 12063
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.283598667160311e-06,
      "loss": 1.2807,
      "step": 12064
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.282117734172529e-06,
      "loss": 0.9446,
      "step": 12065
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2806368011847466e-06,
      "loss": 1.0448,
      "step": 12066
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.279155868196964e-06,
      "loss": 1.2488,
      "step": 12067
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.277674935209182e-06,
      "loss": 1.2895,
      "step": 12068
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2761940022213996e-06,
      "loss": 1.1991,
      "step": 12069
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.274713069233617e-06,
      "loss": 1.1722,
      "step": 12070
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.273232136245835e-06,
      "loss": 1.1911,
      "step": 12071
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2717512032580525e-06,
      "loss": 1.1044,
      "step": 12072
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2702702702702705e-06,
      "loss": 1.3099,
      "step": 12073
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.268789337282488e-06,
      "loss": 1.3025,
      "step": 12074
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.267308404294706e-06,
      "loss": 1.2774,
      "step": 12075
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.265827471306924e-06,
      "loss": 1.0473,
      "step": 12076
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2643465383191414e-06,
      "loss": 1.0347,
      "step": 12077
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.262865605331359e-06,
      "loss": 1.1033,
      "step": 12078
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.261384672343577e-06,
      "loss": 1.145,
      "step": 12079
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2599037393557944e-06,
      "loss": 1.1702,
      "step": 12080
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.258422806368012e-06,
      "loss": 1.1458,
      "step": 12081
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.25694187338023e-06,
      "loss": 1.1223,
      "step": 12082
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2554609403924473e-06,
      "loss": 1.3061,
      "step": 12083
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.253980007404665e-06,
      "loss": 1.3228,
      "step": 12084
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.2524990744168828e-06,
      "loss": 1.3304,
      "step": 12085
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2510181414291003e-06,
      "loss": 1.2412,
      "step": 12086
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2495372084413182e-06,
      "loss": 1.1095,
      "step": 12087
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2480562754535358e-06,
      "loss": 1.309,
      "step": 12088
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2465753424657537e-06,
      "loss": 1.4413,
      "step": 12089
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2450944094779712e-06,
      "loss": 1.0261,
      "step": 12090
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.243613476490189e-06,
      "loss": 1.1769,
      "step": 12091
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2421325435024067e-06,
      "loss": 1.1386,
      "step": 12092
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2406516105146246e-06,
      "loss": 1.3631,
      "step": 12093
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.239170677526842e-06,
      "loss": 1.2068,
      "step": 12094
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2376897445390596e-06,
      "loss": 1.1737,
      "step": 12095
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2362088115512776e-06,
      "loss": 1.1665,
      "step": 12096
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.234727878563495e-06,
      "loss": 1.09,
      "step": 12097
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2332469455757126e-06,
      "loss": 1.0999,
      "step": 12098
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2317660125879306e-06,
      "loss": 1.356,
      "step": 12099
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.230285079600148e-06,
      "loss": 1.2698,
      "step": 12100
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.228804146612366e-06,
      "loss": 1.3514,
      "step": 12101
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2273232136245835e-06,
      "loss": 1.1838,
      "step": 12102
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2258422806368015e-06,
      "loss": 1.1563,
      "step": 12103
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.224361347649019e-06,
      "loss": 1.1371,
      "step": 12104
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.222880414661237e-06,
      "loss": 1.1843,
      "step": 12105
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2213994816734545e-06,
      "loss": 1.0473,
      "step": 12106
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2199185486856724e-06,
      "loss": 1.1755,
      "step": 12107
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.21843761569789e-06,
      "loss": 1.2478,
      "step": 12108
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2169566827101074e-06,
      "loss": 1.3556,
      "step": 12109
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2154757497223254e-06,
      "loss": 1.0745,
      "step": 12110
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.213994816734543e-06,
      "loss": 1.1422,
      "step": 12111
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2125138837467604e-06,
      "loss": 1.0453,
      "step": 12112
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2110329507589783e-06,
      "loss": 1.1977,
      "step": 12113
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.209552017771196e-06,
      "loss": 1.1226,
      "step": 12114
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.208071084783414e-06,
      "loss": 1.2062,
      "step": 12115
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2065901517956313e-06,
      "loss": 1.1868,
      "step": 12116
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2051092188078493e-06,
      "loss": 1.2604,
      "step": 12117
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2036282858200668e-06,
      "loss": 1.1233,
      "step": 12118
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2021473528322847e-06,
      "loss": 1.4067,
      "step": 12119
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.2006664198445022e-06,
      "loss": 1.26,
      "step": 12120
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.19918548685672e-06,
      "loss": 1.1579,
      "step": 12121
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.1977045538689377e-06,
      "loss": 1.3068,
      "step": 12122
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.196223620881155e-06,
      "loss": 1.1917,
      "step": 12123
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.194742687893373e-06,
      "loss": 1.1427,
      "step": 12124
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.1932617549055907e-06,
      "loss": 1.1282,
      "step": 12125
    },
    {
      "epoch": 2.67,
      "eval_loss": NaN,
      "eval_runtime": 4.3787,
      "eval_samples_per_second": 352.618,
      "eval_steps_per_second": 44.077,
      "step": 12125
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.191780821917808e-06,
      "loss": 1.2763,
      "step": 12126
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.190299888930026e-06,
      "loss": 1.116,
      "step": 12127
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.1888189559422436e-06,
      "loss": 1.0397,
      "step": 12128
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.1873380229544616e-06,
      "loss": 1.1,
      "step": 12129
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.185857089966679e-06,
      "loss": 1.1078,
      "step": 12130
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.184376156978897e-06,
      "loss": 1.1721,
      "step": 12131
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1828952239911145e-06,
      "loss": 1.0373,
      "step": 12132
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1814142910033325e-06,
      "loss": 1.4145,
      "step": 12133
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.17993335801555e-06,
      "loss": 1.1122,
      "step": 12134
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.178452425027768e-06,
      "loss": 1.2765,
      "step": 12135
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1769714920399855e-06,
      "loss": 1.2544,
      "step": 12136
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.175490559052203e-06,
      "loss": 1.1573,
      "step": 12137
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.174009626064421e-06,
      "loss": 1.1771,
      "step": 12138
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1725286930766384e-06,
      "loss": 1.233,
      "step": 12139
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.171047760088856e-06,
      "loss": 1.2769,
      "step": 12140
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.169566827101074e-06,
      "loss": 1.4334,
      "step": 12141
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1680858941132914e-06,
      "loss": 1.1709,
      "step": 12142
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.166604961125509e-06,
      "loss": 1.442,
      "step": 12143
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.165124028137727e-06,
      "loss": 1.1345,
      "step": 12144
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1636430951499444e-06,
      "loss": 1.1109,
      "step": 12145
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1621621621621623e-06,
      "loss": 1.2038,
      "step": 12146
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1606812291743803e-06,
      "loss": 0.9198,
      "step": 12147
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1592002961865978e-06,
      "loss": 1.3427,
      "step": 12148
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1577193631988157e-06,
      "loss": 1.2492,
      "step": 12149
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1562384302110332e-06,
      "loss": 1.0863,
      "step": 12150
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1547574972232508e-06,
      "loss": 1.1009,
      "step": 12151
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1532765642354687e-06,
      "loss": 1.2039,
      "step": 12152
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1517956312476862e-06,
      "loss": 1.3216,
      "step": 12153
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1503146982599037e-06,
      "loss": 1.2912,
      "step": 12154
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1488337652721217e-06,
      "loss": 1.259,
      "step": 12155
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.147352832284339e-06,
      "loss": 1.2458,
      "step": 12156
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1458718992965567e-06,
      "loss": 1.3878,
      "step": 12157
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1443909663087746e-06,
      "loss": 1.2408,
      "step": 12158
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.142910033320992e-06,
      "loss": 1.2733,
      "step": 12159
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.14142910033321e-06,
      "loss": 1.1014,
      "step": 12160
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1399481673454276e-06,
      "loss": 1.1423,
      "step": 12161
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1384672343576456e-06,
      "loss": 1.2952,
      "step": 12162
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1369863013698635e-06,
      "loss": 1.1075,
      "step": 12163
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.135505368382081e-06,
      "loss": 1.4055,
      "step": 12164
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1340244353942985e-06,
      "loss": 1.2651,
      "step": 12165
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1325435024065165e-06,
      "loss": 0.9753,
      "step": 12166
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.131062569418734e-06,
      "loss": 1.2649,
      "step": 12167
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1295816364309515e-06,
      "loss": 1.0407,
      "step": 12168
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1281007034431695e-06,
      "loss": 1.3069,
      "step": 12169
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.126619770455387e-06,
      "loss": 0.8816,
      "step": 12170
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1251388374676045e-06,
      "loss": 1.1669,
      "step": 12171
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1236579044798224e-06,
      "loss": 1.1771,
      "step": 12172
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.12217697149204e-06,
      "loss": 1.2616,
      "step": 12173
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.120696038504258e-06,
      "loss": 1.2317,
      "step": 12174
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1192151055164754e-06,
      "loss": 1.2714,
      "step": 12175
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.1177341725286933e-06,
      "loss": 1.2491,
      "step": 12176
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.116253239540911e-06,
      "loss": 1.222,
      "step": 12177
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.114772306553129e-06,
      "loss": 1.0541,
      "step": 12178
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.1132913735653463e-06,
      "loss": 1.5096,
      "step": 12179
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.1118104405775643e-06,
      "loss": 1.1872,
      "step": 12180
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.1103295075897818e-06,
      "loss": 1.2269,
      "step": 12181
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.1088485746019993e-06,
      "loss": 1.16,
      "step": 12182
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.1073676416142172e-06,
      "loss": 1.1234,
      "step": 12183
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.1058867086264347e-06,
      "loss": 1.1266,
      "step": 12184
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.1044057756386523e-06,
      "loss": 1.3303,
      "step": 12185
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.10292484265087e-06,
      "loss": 1.2657,
      "step": 12186
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.1014439096630877e-06,
      "loss": 1.221,
      "step": 12187
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0999629766753057e-06,
      "loss": 1.3071,
      "step": 12188
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.098482043687523e-06,
      "loss": 1.2333,
      "step": 12189
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.097001110699741e-06,
      "loss": 0.9375,
      "step": 12190
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0955201777119586e-06,
      "loss": 1.2998,
      "step": 12191
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0940392447241766e-06,
      "loss": 1.2095,
      "step": 12192
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.092558311736394e-06,
      "loss": 1.161,
      "step": 12193
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.091077378748612e-06,
      "loss": 1.3037,
      "step": 12194
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0895964457608295e-06,
      "loss": 1.2549,
      "step": 12195
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.088115512773047e-06,
      "loss": 1.0724,
      "step": 12196
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.086634579785265e-06,
      "loss": 1.0091,
      "step": 12197
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0851536467974825e-06,
      "loss": 1.1482,
      "step": 12198
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0836727138097e-06,
      "loss": 1.2455,
      "step": 12199
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.082191780821918e-06,
      "loss": 1.0142,
      "step": 12200
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0807108478341355e-06,
      "loss": 1.3403,
      "step": 12201
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0792299148463534e-06,
      "loss": 1.4543,
      "step": 12202
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.077748981858571e-06,
      "loss": 1.2929,
      "step": 12203
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.076268048870789e-06,
      "loss": 1.2422,
      "step": 12204
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0747871158830064e-06,
      "loss": 1.2876,
      "step": 12205
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0733061828952244e-06,
      "loss": 1.3537,
      "step": 12206
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.071825249907442e-06,
      "loss": 1.2691,
      "step": 12207
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.07034431691966e-06,
      "loss": 1.2168,
      "step": 12208
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0688633839318773e-06,
      "loss": 1.1954,
      "step": 12209
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.067382450944095e-06,
      "loss": 1.167,
      "step": 12210
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0659015179563128e-06,
      "loss": 1.0181,
      "step": 12211
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0644205849685303e-06,
      "loss": 1.2058,
      "step": 12212
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.062939651980748e-06,
      "loss": 1.2272,
      "step": 12213
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0614587189929658e-06,
      "loss": 1.3115,
      "step": 12214
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0599777860051833e-06,
      "loss": 1.3438,
      "step": 12215
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0584968530174012e-06,
      "loss": 1.4769,
      "step": 12216
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0570159200296187e-06,
      "loss": 1.2146,
      "step": 12217
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0555349870418362e-06,
      "loss": 1.1586,
      "step": 12218
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.054054054054054e-06,
      "loss": 1.1094,
      "step": 12219
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.052573121066272e-06,
      "loss": 1.3633,
      "step": 12220
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.0510921880784896e-06,
      "loss": 1.2642,
      "step": 12221
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0496112550907076e-06,
      "loss": 1.104,
      "step": 12222
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.048130322102925e-06,
      "loss": 1.1085,
      "step": 12223
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0466493891151426e-06,
      "loss": 1.2209,
      "step": 12224
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0451684561273606e-06,
      "loss": 1.1862,
      "step": 12225
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.043687523139578e-06,
      "loss": 1.3139,
      "step": 12226
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0422065901517956e-06,
      "loss": 1.0659,
      "step": 12227
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0407256571640135e-06,
      "loss": 1.3387,
      "step": 12228
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.039244724176231e-06,
      "loss": 1.1671,
      "step": 12229
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.037763791188449e-06,
      "loss": 1.2528,
      "step": 12230
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0362828582006665e-06,
      "loss": 1.17,
      "step": 12231
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.034801925212884e-06,
      "loss": 1.0859,
      "step": 12232
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.033320992225102e-06,
      "loss": 1.108,
      "step": 12233
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0318400592373195e-06,
      "loss": 1.1478,
      "step": 12234
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0303591262495374e-06,
      "loss": 1.2424,
      "step": 12235
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0288781932617554e-06,
      "loss": 1.2172,
      "step": 12236
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.027397260273973e-06,
      "loss": 1.0683,
      "step": 12237
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0259163272861904e-06,
      "loss": 1.3506,
      "step": 12238
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0244353942984083e-06,
      "loss": 1.1212,
      "step": 12239
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.022954461310626e-06,
      "loss": 1.1865,
      "step": 12240
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0214735283228434e-06,
      "loss": 1.3638,
      "step": 12241
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0199925953350613e-06,
      "loss": 1.2973,
      "step": 12242
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.018511662347279e-06,
      "loss": 1.3276,
      "step": 12243
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0170307293594968e-06,
      "loss": 1.0994,
      "step": 12244
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0155497963717143e-06,
      "loss": 1.0246,
      "step": 12245
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.014068863383932e-06,
      "loss": 1.1466,
      "step": 12246
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0125879303961497e-06,
      "loss": 1.3706,
      "step": 12247
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0111069974083673e-06,
      "loss": 1.0709,
      "step": 12248
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.009626064420585e-06,
      "loss": 1.0441,
      "step": 12249
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0081451314328027e-06,
      "loss": 1.2624,
      "step": 12250
    },
    {
      "epoch": 2.7,
      "eval_loss": NaN,
      "eval_runtime": 4.3802,
      "eval_samples_per_second": 352.497,
      "eval_steps_per_second": 44.062,
      "step": 12250
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0066641984450207e-06,
      "loss": 1.2731,
      "step": 12251
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.005183265457238e-06,
      "loss": 0.9597,
      "step": 12252
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.003702332469456e-06,
      "loss": 1.0617,
      "step": 12253
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.0022213994816736e-06,
      "loss": 1.2819,
      "step": 12254
    },
    {
      "epoch": 2.7,
      "learning_rate": 2.000740466493891e-06,
      "loss": 1.2745,
      "step": 12255
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.999259533506109e-06,
      "loss": 1.119,
      "step": 12256
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.9977786005183266e-06,
      "loss": 1.3943,
      "step": 12257
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.996297667530544e-06,
      "loss": 1.1332,
      "step": 12258
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.994816734542762e-06,
      "loss": 1.1833,
      "step": 12259
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.9933358015549796e-06,
      "loss": 1.198,
      "step": 12260
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.9918548685671975e-06,
      "loss": 1.0244,
      "step": 12261
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.990373935579415e-06,
      "loss": 1.1837,
      "step": 12262
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.988893002591633e-06,
      "loss": 1.0872,
      "step": 12263
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.9874120696038505e-06,
      "loss": 1.0143,
      "step": 12264
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.9859311366160684e-06,
      "loss": 1.1823,
      "step": 12265
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.984450203628286e-06,
      "loss": 1.1584,
      "step": 12266
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.982969270640504e-06,
      "loss": 1.191,
      "step": 12267
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9814883376527214e-06,
      "loss": 0.9636,
      "step": 12268
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.980007404664939e-06,
      "loss": 1.3014,
      "step": 12269
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.978526471677157e-06,
      "loss": 1.2572,
      "step": 12270
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9770455386893744e-06,
      "loss": 1.092,
      "step": 12271
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.975564605701592e-06,
      "loss": 0.9417,
      "step": 12272
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.97408367271381e-06,
      "loss": 1.4249,
      "step": 12273
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9726027397260274e-06,
      "loss": 1.2019,
      "step": 12274
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9711218067382453e-06,
      "loss": 1.2827,
      "step": 12275
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.969640873750463e-06,
      "loss": 1.1848,
      "step": 12276
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9681599407626808e-06,
      "loss": 1.2301,
      "step": 12277
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9666790077748983e-06,
      "loss": 1.2151,
      "step": 12278
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9651980747871162e-06,
      "loss": 1.1564,
      "step": 12279
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9637171417993337e-06,
      "loss": 1.1722,
      "step": 12280
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9622362088115517e-06,
      "loss": 1.1581,
      "step": 12281
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.960755275823769e-06,
      "loss": 1.3196,
      "step": 12282
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9592743428359867e-06,
      "loss": 1.1716,
      "step": 12283
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9577934098482046e-06,
      "loss": 1.0793,
      "step": 12284
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.956312476860422e-06,
      "loss": 1.1482,
      "step": 12285
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9548315438726397e-06,
      "loss": 1.1591,
      "step": 12286
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9533506108848576e-06,
      "loss": 1.2817,
      "step": 12287
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.951869677897075e-06,
      "loss": 1.3851,
      "step": 12288
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.950388744909293e-06,
      "loss": 1.1987,
      "step": 12289
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9489078119215106e-06,
      "loss": 1.5208,
      "step": 12290
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9474268789337285e-06,
      "loss": 1.2129,
      "step": 12291
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.945945945945946e-06,
      "loss": 1.3107,
      "step": 12292
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.944465012958164e-06,
      "loss": 1.1235,
      "step": 12293
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9429840799703815e-06,
      "loss": 1.3629,
      "step": 12294
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9415031469825995e-06,
      "loss": 1.3368,
      "step": 12295
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.940022213994817e-06,
      "loss": 1.1498,
      "step": 12296
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9385412810070345e-06,
      "loss": 1.3153,
      "step": 12297
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9370603480192524e-06,
      "loss": 1.1649,
      "step": 12298
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.93557941503147e-06,
      "loss": 1.4521,
      "step": 12299
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9340984820436875e-06,
      "loss": 0.9997,
      "step": 12300
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9326175490559054e-06,
      "loss": 1.2564,
      "step": 12301
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.931136616068123e-06,
      "loss": 1.1376,
      "step": 12302
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.929655683080341e-06,
      "loss": 1.0903,
      "step": 12303
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9281747500925584e-06,
      "loss": 1.2454,
      "step": 12304
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.926693817104776e-06,
      "loss": 1.3034,
      "step": 12305
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.925212884116994e-06,
      "loss": 1.173,
      "step": 12306
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9237319511292118e-06,
      "loss": 1.1502,
      "step": 12307
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9222510181414293e-06,
      "loss": 1.0948,
      "step": 12308
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9207700851536472e-06,
      "loss": 1.0834,
      "step": 12309
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9192891521658647e-06,
      "loss": 1.1892,
      "step": 12310
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9178082191780823e-06,
      "loss": 1.1969,
      "step": 12311
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.9163272861903e-06,
      "loss": 1.1437,
      "step": 12312
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9148463532025177e-06,
      "loss": 1.0715,
      "step": 12313
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9133654202147352e-06,
      "loss": 1.0418,
      "step": 12314
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.911884487226953e-06,
      "loss": 1.2086,
      "step": 12315
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9104035542391707e-06,
      "loss": 1.1367,
      "step": 12316
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9089226212513886e-06,
      "loss": 1.4374,
      "step": 12317
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.907441688263606e-06,
      "loss": 1.1028,
      "step": 12318
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9059607552758239e-06,
      "loss": 1.0695,
      "step": 12319
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9044798222880418e-06,
      "loss": 1.5716,
      "step": 12320
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9029988893002593e-06,
      "loss": 1.3607,
      "step": 12321
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9015179563124769e-06,
      "loss": 1.4945,
      "step": 12322
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9000370233246948e-06,
      "loss": 1.1046,
      "step": 12323
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8985560903369123e-06,
      "loss": 1.0434,
      "step": 12324
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.89707515734913e-06,
      "loss": 1.198,
      "step": 12325
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.895594224361348e-06,
      "loss": 1.1768,
      "step": 12326
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8941132913735655e-06,
      "loss": 1.2626,
      "step": 12327
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.892632358385783e-06,
      "loss": 1.1901,
      "step": 12328
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.891151425398001e-06,
      "loss": 1.2314,
      "step": 12329
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8896704924102185e-06,
      "loss": 1.1635,
      "step": 12330
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8881895594224364e-06,
      "loss": 1.0978,
      "step": 12331
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.886708626434654e-06,
      "loss": 1.1138,
      "step": 12332
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8852276934468717e-06,
      "loss": 1.4479,
      "step": 12333
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8837467604590896e-06,
      "loss": 1.2142,
      "step": 12334
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8822658274713071e-06,
      "loss": 1.228,
      "step": 12335
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8807848944835246e-06,
      "loss": 1.1224,
      "step": 12336
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8793039614957426e-06,
      "loss": 1.2808,
      "step": 12337
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.87782302850796e-06,
      "loss": 1.1444,
      "step": 12338
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8763420955201778e-06,
      "loss": 0.9579,
      "step": 12339
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8748611625323955e-06,
      "loss": 1.0255,
      "step": 12340
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8733802295446133e-06,
      "loss": 1.1262,
      "step": 12341
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8718992965568308e-06,
      "loss": 1.3998,
      "step": 12342
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8704183635690487e-06,
      "loss": 1.1754,
      "step": 12343
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8689374305812662e-06,
      "loss": 1.2688,
      "step": 12344
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8674564975934842e-06,
      "loss": 1.0252,
      "step": 12345
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8659755646057017e-06,
      "loss": 1.1206,
      "step": 12346
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8644946316179194e-06,
      "loss": 1.0903,
      "step": 12347
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8630136986301372e-06,
      "loss": 1.0539,
      "step": 12348
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8615327656423549e-06,
      "loss": 1.1311,
      "step": 12349
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8600518326545724e-06,
      "loss": 1.1641,
      "step": 12350
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8585708996667903e-06,
      "loss": 1.1218,
      "step": 12351
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8570899666790079e-06,
      "loss": 1.1454,
      "step": 12352
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8556090336912256e-06,
      "loss": 1.2657,
      "step": 12353
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8541281007034433e-06,
      "loss": 1.4255,
      "step": 12354
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.852647167715661e-06,
      "loss": 1.2433,
      "step": 12355
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8511662347278786e-06,
      "loss": 1.1573,
      "step": 12356
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.8496853017400965e-06,
      "loss": 1.176,
      "step": 12357
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.848204368752314e-06,
      "loss": 1.2187,
      "step": 12358
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8467234357645318e-06,
      "loss": 1.1215,
      "step": 12359
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8452425027767495e-06,
      "loss": 1.2137,
      "step": 12360
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8437615697889672e-06,
      "loss": 1.3427,
      "step": 12361
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.842280636801185e-06,
      "loss": 1.1485,
      "step": 12362
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8407997038134027e-06,
      "loss": 1.0781,
      "step": 12363
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8393187708256202e-06,
      "loss": 1.1499,
      "step": 12364
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8378378378378381e-06,
      "loss": 1.2048,
      "step": 12365
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8363569048500556e-06,
      "loss": 1.1559,
      "step": 12366
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8348759718622734e-06,
      "loss": 1.0638,
      "step": 12367
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.833395038874491e-06,
      "loss": 0.9912,
      "step": 12368
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8319141058867088e-06,
      "loss": 1.3053,
      "step": 12369
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8304331728989263e-06,
      "loss": 1.0342,
      "step": 12370
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8289522399111443e-06,
      "loss": 1.3498,
      "step": 12371
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8274713069233618e-06,
      "loss": 1.1241,
      "step": 12372
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8259903739355795e-06,
      "loss": 1.3116,
      "step": 12373
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8245094409477973e-06,
      "loss": 1.2253,
      "step": 12374
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.823028507960015e-06,
      "loss": 0.976,
      "step": 12375
    },
    {
      "epoch": 2.73,
      "eval_loss": NaN,
      "eval_runtime": 4.376,
      "eval_samples_per_second": 352.831,
      "eval_steps_per_second": 44.104,
      "step": 12375
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8215475749722327e-06,
      "loss": 1.1109,
      "step": 12376
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8200666419844504e-06,
      "loss": 1.378,
      "step": 12377
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.818585708996668e-06,
      "loss": 1.1227,
      "step": 12378
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.817104776008886e-06,
      "loss": 1.3187,
      "step": 12379
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8156238430211034e-06,
      "loss": 1.3916,
      "step": 12380
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8141429100333211e-06,
      "loss": 1.1636,
      "step": 12381
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8126619770455389e-06,
      "loss": 1.1683,
      "step": 12382
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8111810440577566e-06,
      "loss": 1.3615,
      "step": 12383
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8097001110699741e-06,
      "loss": 1.3585,
      "step": 12384
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.808219178082192e-06,
      "loss": 1.4192,
      "step": 12385
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8067382450944096e-06,
      "loss": 1.1161,
      "step": 12386
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.805257312106627e-06,
      "loss": 1.0817,
      "step": 12387
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.803776379118845e-06,
      "loss": 1.1662,
      "step": 12388
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8022954461310628e-06,
      "loss": 1.253,
      "step": 12389
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.8008145131432805e-06,
      "loss": 1.1426,
      "step": 12390
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7993335801554982e-06,
      "loss": 1.3271,
      "step": 12391
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7978526471677157e-06,
      "loss": 1.0156,
      "step": 12392
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7963717141799337e-06,
      "loss": 0.9763,
      "step": 12393
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7948907811921512e-06,
      "loss": 1.1831,
      "step": 12394
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7934098482043687e-06,
      "loss": 1.2825,
      "step": 12395
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7919289152165867e-06,
      "loss": 1.4727,
      "step": 12396
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7904479822288044e-06,
      "loss": 1.172,
      "step": 12397
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.788967049241022e-06,
      "loss": 1.1706,
      "step": 12398
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7874861162532398e-06,
      "loss": 1.1308,
      "step": 12399
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7860051832654574e-06,
      "loss": 1.2426,
      "step": 12400
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7845242502776749e-06,
      "loss": 1.154,
      "step": 12401
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7830433172898928e-06,
      "loss": 1.5015,
      "step": 12402
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.7815623843021103e-06,
      "loss": 1.1914,
      "step": 12403
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7800814513143283e-06,
      "loss": 1.0822,
      "step": 12404
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.778600518326546e-06,
      "loss": 1.4503,
      "step": 12405
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7771195853387635e-06,
      "loss": 1.0875,
      "step": 12406
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7756386523509815e-06,
      "loss": 1.2939,
      "step": 12407
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.774157719363199e-06,
      "loss": 1.0921,
      "step": 12408
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7726767863754165e-06,
      "loss": 1.2866,
      "step": 12409
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7711958533876344e-06,
      "loss": 0.9605,
      "step": 12410
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.769714920399852e-06,
      "loss": 1.4869,
      "step": 12411
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7682339874120697e-06,
      "loss": 1.1441,
      "step": 12412
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7667530544242876e-06,
      "loss": 1.2351,
      "step": 12413
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7652721214365051e-06,
      "loss": 1.1323,
      "step": 12414
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7637911884487227e-06,
      "loss": 1.1429,
      "step": 12415
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7623102554609406e-06,
      "loss": 1.1829,
      "step": 12416
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7608293224731581e-06,
      "loss": 1.2864,
      "step": 12417
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.759348389485376e-06,
      "loss": 1.1342,
      "step": 12418
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7578674564975936e-06,
      "loss": 1.1887,
      "step": 12419
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7563865235098113e-06,
      "loss": 1.4749,
      "step": 12420
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7549055905220292e-06,
      "loss": 1.2013,
      "step": 12421
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7534246575342468e-06,
      "loss": 1.3461,
      "step": 12422
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7519437245464643e-06,
      "loss": 1.0565,
      "step": 12423
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7504627915586822e-06,
      "loss": 1.0685,
      "step": 12424
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7489818585708997e-06,
      "loss": 1.1228,
      "step": 12425
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7475009255831175e-06,
      "loss": 1.0774,
      "step": 12426
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7460199925953352e-06,
      "loss": 1.3276,
      "step": 12427
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.744539059607553e-06,
      "loss": 1.3493,
      "step": 12428
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7430581266197704e-06,
      "loss": 1.319,
      "step": 12429
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7415771936319884e-06,
      "loss": 1.2997,
      "step": 12430
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7400962606442059e-06,
      "loss": 1.2316,
      "step": 12431
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7386153276564238e-06,
      "loss": 1.3632,
      "step": 12432
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7371343946686413e-06,
      "loss": 1.1269,
      "step": 12433
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.735653461680859e-06,
      "loss": 1.3936,
      "step": 12434
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7341725286930768e-06,
      "loss": 1.4374,
      "step": 12435
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7326915957052945e-06,
      "loss": 1.1219,
      "step": 12436
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.731210662717512e-06,
      "loss": 1.0583,
      "step": 12437
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.72972972972973e-06,
      "loss": 1.2837,
      "step": 12438
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7282487967419475e-06,
      "loss": 1.158,
      "step": 12439
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7267678637541652e-06,
      "loss": 1.239,
      "step": 12440
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.725286930766383e-06,
      "loss": 1.2342,
      "step": 12441
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7238059977786007e-06,
      "loss": 1.081,
      "step": 12442
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7223250647908182e-06,
      "loss": 1.4631,
      "step": 12443
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7208441318030361e-06,
      "loss": 1.1912,
      "step": 12444
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7193631988152537e-06,
      "loss": 1.2033,
      "step": 12445
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7178822658274716e-06,
      "loss": 1.1116,
      "step": 12446
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7164013328396891e-06,
      "loss": 1.1301,
      "step": 12447
    },
    {
      "epoch": 2.74,
      "learning_rate": 1.7149203998519069e-06,
      "loss": 1.0221,
      "step": 12448
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.7134394668641246e-06,
      "loss": 1.2641,
      "step": 12449
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.7119585338763423e-06,
      "loss": 1.1448,
      "step": 12450
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.7104776008885598e-06,
      "loss": 1.3595,
      "step": 12451
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.7089966679007778e-06,
      "loss": 0.9047,
      "step": 12452
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.7075157349129953e-06,
      "loss": 1.1002,
      "step": 12453
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.706034801925213e-06,
      "loss": 1.1,
      "step": 12454
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.7045538689374307e-06,
      "loss": 1.301,
      "step": 12455
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.7030729359496485e-06,
      "loss": 1.1381,
      "step": 12456
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.701592002961866e-06,
      "loss": 1.1951,
      "step": 12457
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.700111069974084e-06,
      "loss": 1.1238,
      "step": 12458
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6986301369863014e-06,
      "loss": 1.3005,
      "step": 12459
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6971492039985192e-06,
      "loss": 1.1839,
      "step": 12460
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.695668271010737e-06,
      "loss": 1.0605,
      "step": 12461
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6941873380229546e-06,
      "loss": 1.1137,
      "step": 12462
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6927064050351724e-06,
      "loss": 1.1487,
      "step": 12463
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.69122547204739e-06,
      "loss": 1.1207,
      "step": 12464
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6897445390596076e-06,
      "loss": 1.2741,
      "step": 12465
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6882636060718255e-06,
      "loss": 1.4841,
      "step": 12466
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.686782673084043e-06,
      "loss": 1.3063,
      "step": 12467
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6853017400962608e-06,
      "loss": 1.1255,
      "step": 12468
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6838208071084785e-06,
      "loss": 0.9633,
      "step": 12469
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6823398741206962e-06,
      "loss": 1.205,
      "step": 12470
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6808589411329138e-06,
      "loss": 1.4282,
      "step": 12471
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6793780081451317e-06,
      "loss": 1.2642,
      "step": 12472
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6778970751573492e-06,
      "loss": 1.0755,
      "step": 12473
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6764161421695667e-06,
      "loss": 1.4719,
      "step": 12474
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6749352091817847e-06,
      "loss": 1.4818,
      "step": 12475
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6734542761940024e-06,
      "loss": 1.1975,
      "step": 12476
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6719733432062201e-06,
      "loss": 1.2989,
      "step": 12477
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6704924102184379e-06,
      "loss": 1.1517,
      "step": 12478
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6690114772306554e-06,
      "loss": 1.0359,
      "step": 12479
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6675305442428733e-06,
      "loss": 1.1513,
      "step": 12480
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6660496112550908e-06,
      "loss": 1.2224,
      "step": 12481
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6645686782673084e-06,
      "loss": 1.0772,
      "step": 12482
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6630877452795263e-06,
      "loss": 1.291,
      "step": 12483
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.661606812291744e-06,
      "loss": 1.1858,
      "step": 12484
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6601258793039615e-06,
      "loss": 1.2672,
      "step": 12485
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6586449463161795e-06,
      "loss": 1.3538,
      "step": 12486
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.657164013328397e-06,
      "loss": 0.9212,
      "step": 12487
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6556830803406145e-06,
      "loss": 0.9883,
      "step": 12488
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6542021473528325e-06,
      "loss": 0.9772,
      "step": 12489
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.65272121436505e-06,
      "loss": 1.2176,
      "step": 12490
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.651240281377268e-06,
      "loss": 1.2203,
      "step": 12491
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6497593483894856e-06,
      "loss": 1.2081,
      "step": 12492
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.6482784154017032e-06,
      "loss": 1.1617,
      "step": 12493
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.646797482413921e-06,
      "loss": 1.0698,
      "step": 12494
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6453165494261386e-06,
      "loss": 1.2171,
      "step": 12495
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6438356164383561e-06,
      "loss": 1.3514,
      "step": 12496
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.642354683450574e-06,
      "loss": 1.0958,
      "step": 12497
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6408737504627916e-06,
      "loss": 1.2943,
      "step": 12498
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6393928174750093e-06,
      "loss": 1.2453,
      "step": 12499
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6379118844872273e-06,
      "loss": 1.2381,
      "step": 12500
    },
    {
      "epoch": 2.76,
      "eval_loss": NaN,
      "eval_runtime": 4.3733,
      "eval_samples_per_second": 353.049,
      "eval_steps_per_second": 44.131,
      "step": 12500
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6364309514994448e-06,
      "loss": 1.1984,
      "step": 12501
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6349500185116623e-06,
      "loss": 1.1478,
      "step": 12502
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6334690855238802e-06,
      "loss": 1.3449,
      "step": 12503
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6319881525360978e-06,
      "loss": 1.2481,
      "step": 12504
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6305072195483157e-06,
      "loss": 1.3117,
      "step": 12505
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6290262865605332e-06,
      "loss": 1.3993,
      "step": 12506
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.627545353572751e-06,
      "loss": 1.24,
      "step": 12507
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6260644205849689e-06,
      "loss": 1.115,
      "step": 12508
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6245834875971864e-06,
      "loss": 1.0106,
      "step": 12509
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.623102554609404e-06,
      "loss": 1.1238,
      "step": 12510
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6216216216216219e-06,
      "loss": 1.2909,
      "step": 12511
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6201406886338394e-06,
      "loss": 1.1362,
      "step": 12512
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.618659755646057e-06,
      "loss": 1.4895,
      "step": 12513
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6171788226582748e-06,
      "loss": 1.2371,
      "step": 12514
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6156978896704926e-06,
      "loss": 1.2443,
      "step": 12515
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.61421695668271e-06,
      "loss": 1.2813,
      "step": 12516
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.612736023694928e-06,
      "loss": 0.9748,
      "step": 12517
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6112550907071455e-06,
      "loss": 1.1807,
      "step": 12518
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6097741577193635e-06,
      "loss": 1.5036,
      "step": 12519
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.608293224731581e-06,
      "loss": 1.3248,
      "step": 12520
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6068122917437987e-06,
      "loss": 1.1929,
      "step": 12521
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6053313587560164e-06,
      "loss": 1.2884,
      "step": 12522
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6038504257682342e-06,
      "loss": 1.1106,
      "step": 12523
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6023694927804517e-06,
      "loss": 1.5112,
      "step": 12524
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.6008885597926696e-06,
      "loss": 1.3703,
      "step": 12525
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5994076268048871e-06,
      "loss": 1.1432,
      "step": 12526
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5979266938171049e-06,
      "loss": 1.3335,
      "step": 12527
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5964457608293226e-06,
      "loss": 1.2554,
      "step": 12528
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5949648278415403e-06,
      "loss": 1.295,
      "step": 12529
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5934838948537578e-06,
      "loss": 1.2981,
      "step": 12530
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5920029618659758e-06,
      "loss": 1.1036,
      "step": 12531
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5905220288781933e-06,
      "loss": 1.0905,
      "step": 12532
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5890410958904112e-06,
      "loss": 1.2201,
      "step": 12533
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5875601629026288e-06,
      "loss": 1.2172,
      "step": 12534
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5860792299148465e-06,
      "loss": 1.1015,
      "step": 12535
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5845982969270642e-06,
      "loss": 1.3241,
      "step": 12536
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.583117363939282e-06,
      "loss": 1.0221,
      "step": 12537
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5816364309514995e-06,
      "loss": 1.1577,
      "step": 12538
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.5801554979637174e-06,
      "loss": 1.1313,
      "step": 12539
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.578674564975935e-06,
      "loss": 1.1587,
      "step": 12540
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5771936319881527e-06,
      "loss": 1.2367,
      "step": 12541
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5757126990003704e-06,
      "loss": 1.3366,
      "step": 12542
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5742317660125881e-06,
      "loss": 1.1987,
      "step": 12543
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5727508330248056e-06,
      "loss": 1.1408,
      "step": 12544
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5712699000370236e-06,
      "loss": 1.0531,
      "step": 12545
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.569788967049241e-06,
      "loss": 1.0532,
      "step": 12546
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.568308034061459e-06,
      "loss": 1.1392,
      "step": 12547
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5668271010736765e-06,
      "loss": 1.236,
      "step": 12548
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5653461680858943e-06,
      "loss": 1.1368,
      "step": 12549
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.563865235098112e-06,
      "loss": 1.0721,
      "step": 12550
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5623843021103297e-06,
      "loss": 1.2256,
      "step": 12551
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5609033691225472e-06,
      "loss": 1.388,
      "step": 12552
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5594224361347652e-06,
      "loss": 1.2488,
      "step": 12553
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5579415031469827e-06,
      "loss": 1.0571,
      "step": 12554
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5564605701592002e-06,
      "loss": 1.043,
      "step": 12555
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5549796371714182e-06,
      "loss": 1.2924,
      "step": 12556
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5534987041836359e-06,
      "loss": 1.1738,
      "step": 12557
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5520177711958534e-06,
      "loss": 1.5127,
      "step": 12558
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5505368382080713e-06,
      "loss": 1.2269,
      "step": 12559
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5490559052202889e-06,
      "loss": 1.2465,
      "step": 12560
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5475749722325064e-06,
      "loss": 1.2369,
      "step": 12561
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5460940392447243e-06,
      "loss": 1.1883,
      "step": 12562
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5446131062569418e-06,
      "loss": 1.3342,
      "step": 12563
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5431321732691598e-06,
      "loss": 1.0667,
      "step": 12564
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5416512402813775e-06,
      "loss": 1.202,
      "step": 12565
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.540170307293595e-06,
      "loss": 1.3744,
      "step": 12566
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.538689374305813e-06,
      "loss": 1.1723,
      "step": 12567
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5372084413180305e-06,
      "loss": 1.5248,
      "step": 12568
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.535727508330248e-06,
      "loss": 1.2554,
      "step": 12569
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.534246575342466e-06,
      "loss": 1.1536,
      "step": 12570
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5327656423546835e-06,
      "loss": 1.2014,
      "step": 12571
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5312847093669012e-06,
      "loss": 1.1616,
      "step": 12572
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5298037763791191e-06,
      "loss": 1.1905,
      "step": 12573
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5283228433913366e-06,
      "loss": 1.1578,
      "step": 12574
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5268419104035542e-06,
      "loss": 1.1752,
      "step": 12575
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.525360977415772e-06,
      "loss": 1.344,
      "step": 12576
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5238800444279896e-06,
      "loss": 1.2038,
      "step": 12577
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5223991114402076e-06,
      "loss": 1.075,
      "step": 12578
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.520918178452425e-06,
      "loss": 1.3857,
      "step": 12579
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5194372454646428e-06,
      "loss": 1.2549,
      "step": 12580
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5179563124768607e-06,
      "loss": 1.2024,
      "step": 12581
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5164753794890783e-06,
      "loss": 1.2567,
      "step": 12582
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5149944465012958e-06,
      "loss": 1.2954,
      "step": 12583
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.5135135135135137e-06,
      "loss": 1.2064,
      "step": 12584
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.5120325805257312e-06,
      "loss": 1.3157,
      "step": 12585
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.510551647537949e-06,
      "loss": 1.1195,
      "step": 12586
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.5090707145501667e-06,
      "loss": 1.1171,
      "step": 12587
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.5075897815623844e-06,
      "loss": 1.0737,
      "step": 12588
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.506108848574602e-06,
      "loss": 1.1639,
      "step": 12589
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.5046279155868199e-06,
      "loss": 1.0977,
      "step": 12590
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.5031469825990374e-06,
      "loss": 1.0607,
      "step": 12591
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.5016660496112553e-06,
      "loss": 1.3501,
      "step": 12592
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.5001851166234728e-06,
      "loss": 1.3763,
      "step": 12593
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4987041836356906e-06,
      "loss": 1.361,
      "step": 12594
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4972232506479083e-06,
      "loss": 1.2002,
      "step": 12595
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.495742317660126e-06,
      "loss": 1.1157,
      "step": 12596
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4942613846723436e-06,
      "loss": 1.2885,
      "step": 12597
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4927804516845615e-06,
      "loss": 1.0989,
      "step": 12598
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.491299518696779e-06,
      "loss": 1.2819,
      "step": 12599
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4898185857089967e-06,
      "loss": 1.1003,
      "step": 12600
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4883376527212145e-06,
      "loss": 1.3258,
      "step": 12601
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4868567197334322e-06,
      "loss": 1.2841,
      "step": 12602
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4853757867456497e-06,
      "loss": 1.2505,
      "step": 12603
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4838948537578677e-06,
      "loss": 1.07,
      "step": 12604
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4824139207700852e-06,
      "loss": 1.1776,
      "step": 12605
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4809329877823031e-06,
      "loss": 1.5983,
      "step": 12606
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4794520547945206e-06,
      "loss": 1.2682,
      "step": 12607
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4779711218067384e-06,
      "loss": 1.3164,
      "step": 12608
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.476490188818956e-06,
      "loss": 1.4139,
      "step": 12609
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4750092558311738e-06,
      "loss": 1.2291,
      "step": 12610
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4735283228433913e-06,
      "loss": 1.1912,
      "step": 12611
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4720473898556093e-06,
      "loss": 1.1426,
      "step": 12612
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4705664568678268e-06,
      "loss": 0.8881,
      "step": 12613
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4690855238800445e-06,
      "loss": 1.0493,
      "step": 12614
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4676045908922622e-06,
      "loss": 1.2293,
      "step": 12615
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.46612365790448e-06,
      "loss": 1.2305,
      "step": 12616
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4646427249166975e-06,
      "loss": 1.3839,
      "step": 12617
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4631617919289154e-06,
      "loss": 1.3431,
      "step": 12618
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.461680858941133e-06,
      "loss": 1.1479,
      "step": 12619
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4601999259533509e-06,
      "loss": 1.1965,
      "step": 12620
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4587189929655684e-06,
      "loss": 1.2139,
      "step": 12621
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4572380599777861e-06,
      "loss": 1.1767,
      "step": 12622
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4557571269900039e-06,
      "loss": 1.041,
      "step": 12623
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4542761940022216e-06,
      "loss": 1.3002,
      "step": 12624
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4527952610144391e-06,
      "loss": 1.139,
      "step": 12625
    },
    {
      "epoch": 2.78,
      "eval_loss": NaN,
      "eval_runtime": 4.374,
      "eval_samples_per_second": 352.997,
      "eval_steps_per_second": 44.125,
      "step": 12625
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.451314328026657e-06,
      "loss": 0.9202,
      "step": 12626
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4498333950388746e-06,
      "loss": 1.4992,
      "step": 12627
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.4483524620510923e-06,
      "loss": 1.0483,
      "step": 12628
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.44687152906331e-06,
      "loss": 1.1687,
      "step": 12629
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4453905960755277e-06,
      "loss": 1.0665,
      "step": 12630
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4439096630877453e-06,
      "loss": 1.3421,
      "step": 12631
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4424287300999632e-06,
      "loss": 1.2755,
      "step": 12632
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4409477971121807e-06,
      "loss": 1.399,
      "step": 12633
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4394668641243987e-06,
      "loss": 1.2526,
      "step": 12634
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4379859311366162e-06,
      "loss": 1.1191,
      "step": 12635
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.436504998148834e-06,
      "loss": 1.0879,
      "step": 12636
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4350240651610516e-06,
      "loss": 0.9543,
      "step": 12637
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4335431321732694e-06,
      "loss": 1.3378,
      "step": 12638
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4320621991854869e-06,
      "loss": 1.2368,
      "step": 12639
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4305812661977048e-06,
      "loss": 1.3744,
      "step": 12640
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4291003332099223e-06,
      "loss": 1.1928,
      "step": 12641
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4276194002221399e-06,
      "loss": 1.221,
      "step": 12642
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4261384672343578e-06,
      "loss": 1.2816,
      "step": 12643
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4246575342465755e-06,
      "loss": 1.3448,
      "step": 12644
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.423176601258793e-06,
      "loss": 1.0911,
      "step": 12645
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.421695668271011e-06,
      "loss": 1.2045,
      "step": 12646
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4202147352832285e-06,
      "loss": 1.1781,
      "step": 12647
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4187338022954464e-06,
      "loss": 1.0411,
      "step": 12648
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.417252869307664e-06,
      "loss": 1.323,
      "step": 12649
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4157719363198815e-06,
      "loss": 1.433,
      "step": 12650
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4142910033320994e-06,
      "loss": 1.104,
      "step": 12651
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4128100703443171e-06,
      "loss": 1.427,
      "step": 12652
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4113291373565347e-06,
      "loss": 1.348,
      "step": 12653
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4098482043687526e-06,
      "loss": 1.177,
      "step": 12654
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4083672713809701e-06,
      "loss": 1.0083,
      "step": 12655
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4068863383931876e-06,
      "loss": 1.007,
      "step": 12656
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4054054054054056e-06,
      "loss": 1.1235,
      "step": 12657
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.403924472417623e-06,
      "loss": 1.0993,
      "step": 12658
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4024435394298408e-06,
      "loss": 1.0593,
      "step": 12659
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.4009626064420588e-06,
      "loss": 1.2795,
      "step": 12660
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3994816734542763e-06,
      "loss": 1.2144,
      "step": 12661
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3980007404664942e-06,
      "loss": 1.2336,
      "step": 12662
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3965198074787117e-06,
      "loss": 1.445,
      "step": 12663
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3950388744909293e-06,
      "loss": 1.3804,
      "step": 12664
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3935579415031472e-06,
      "loss": 1.0501,
      "step": 12665
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3920770085153647e-06,
      "loss": 1.1771,
      "step": 12666
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3905960755275824e-06,
      "loss": 1.2601,
      "step": 12667
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3891151425398004e-06,
      "loss": 1.2344,
      "step": 12668
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.387634209552018e-06,
      "loss": 1.3825,
      "step": 12669
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3861532765642354e-06,
      "loss": 1.1388,
      "step": 12670
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3846723435764534e-06,
      "loss": 1.0488,
      "step": 12671
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3831914105886709e-06,
      "loss": 1.1281,
      "step": 12672
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3817104776008886e-06,
      "loss": 1.2893,
      "step": 12673
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.3802295446131063e-06,
      "loss": 0.9221,
      "step": 12674
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.378748611625324e-06,
      "loss": 1.3056,
      "step": 12675
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3772676786375416e-06,
      "loss": 1.2338,
      "step": 12676
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3757867456497595e-06,
      "loss": 1.2049,
      "step": 12677
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.374305812661977e-06,
      "loss": 1.1117,
      "step": 12678
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.372824879674195e-06,
      "loss": 1.1177,
      "step": 12679
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3713439466864125e-06,
      "loss": 1.3672,
      "step": 12680
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3698630136986302e-06,
      "loss": 1.143,
      "step": 12681
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.368382080710848e-06,
      "loss": 1.1786,
      "step": 12682
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3669011477230657e-06,
      "loss": 1.1276,
      "step": 12683
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3654202147352832e-06,
      "loss": 1.1397,
      "step": 12684
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3639392817475011e-06,
      "loss": 1.2624,
      "step": 12685
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3624583487597186e-06,
      "loss": 1.2652,
      "step": 12686
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3609774157719364e-06,
      "loss": 1.2762,
      "step": 12687
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3594964827841541e-06,
      "loss": 1.2462,
      "step": 12688
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3580155497963718e-06,
      "loss": 1.1415,
      "step": 12689
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3565346168085894e-06,
      "loss": 1.3439,
      "step": 12690
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3550536838208073e-06,
      "loss": 1.293,
      "step": 12691
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3535727508330248e-06,
      "loss": 1.2684,
      "step": 12692
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3520918178452427e-06,
      "loss": 1.1568,
      "step": 12693
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3506108848574603e-06,
      "loss": 1.1844,
      "step": 12694
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.349129951869678e-06,
      "loss": 1.2748,
      "step": 12695
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3476490188818957e-06,
      "loss": 1.2739,
      "step": 12696
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3461680858941135e-06,
      "loss": 1.2966,
      "step": 12697
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.344687152906331e-06,
      "loss": 1.298,
      "step": 12698
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.343206219918549e-06,
      "loss": 1.4653,
      "step": 12699
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3417252869307664e-06,
      "loss": 1.1366,
      "step": 12700
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3402443539429842e-06,
      "loss": 1.2602,
      "step": 12701
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3387634209552019e-06,
      "loss": 1.2464,
      "step": 12702
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3372824879674196e-06,
      "loss": 1.0932,
      "step": 12703
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3358015549796371e-06,
      "loss": 1.2036,
      "step": 12704
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.334320621991855e-06,
      "loss": 1.158,
      "step": 12705
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3328396890040726e-06,
      "loss": 1.0444,
      "step": 12706
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3313587560162905e-06,
      "loss": 1.1686,
      "step": 12707
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.329877823028508e-06,
      "loss": 1.0851,
      "step": 12708
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3283968900407258e-06,
      "loss": 1.5191,
      "step": 12709
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3269159570529435e-06,
      "loss": 1.319,
      "step": 12710
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3254350240651612e-06,
      "loss": 1.061,
      "step": 12711
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3239540910773787e-06,
      "loss": 1.1317,
      "step": 12712
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3224731580895967e-06,
      "loss": 1.1703,
      "step": 12713
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3209922251018142e-06,
      "loss": 1.0864,
      "step": 12714
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.319511292114032e-06,
      "loss": 1.0756,
      "step": 12715
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3180303591262497e-06,
      "loss": 1.5252,
      "step": 12716
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3165494261384674e-06,
      "loss": 1.1517,
      "step": 12717
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.315068493150685e-06,
      "loss": 1.2764,
      "step": 12718
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3135875601629028e-06,
      "loss": 1.3106,
      "step": 12719
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3121066271751204e-06,
      "loss": 1.2974,
      "step": 12720
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.3106256941873383e-06,
      "loss": 1.3752,
      "step": 12721
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.3091447611995558e-06,
      "loss": 1.2237,
      "step": 12722
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.3076638282117736e-06,
      "loss": 0.9918,
      "step": 12723
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.3061828952239913e-06,
      "loss": 1.1719,
      "step": 12724
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.304701962236209e-06,
      "loss": 1.0403,
      "step": 12725
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.3032210292484265e-06,
      "loss": 1.148,
      "step": 12726
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.3017400962606445e-06,
      "loss": 1.1204,
      "step": 12727
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.300259163272862e-06,
      "loss": 1.32,
      "step": 12728
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2987782302850795e-06,
      "loss": 1.0703,
      "step": 12729
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2972972972972974e-06,
      "loss": 1.1478,
      "step": 12730
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2958163643095152e-06,
      "loss": 1.2362,
      "step": 12731
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2943354313217327e-06,
      "loss": 1.2093,
      "step": 12732
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2928544983339506e-06,
      "loss": 1.3366,
      "step": 12733
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2913735653461681e-06,
      "loss": 1.2588,
      "step": 12734
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.289892632358386e-06,
      "loss": 1.1258,
      "step": 12735
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2884116993706036e-06,
      "loss": 1.0961,
      "step": 12736
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2869307663828211e-06,
      "loss": 1.2237,
      "step": 12737
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.285449833395039e-06,
      "loss": 1.4635,
      "step": 12738
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2839689004072568e-06,
      "loss": 1.2067,
      "step": 12739
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2824879674194743e-06,
      "loss": 1.4005,
      "step": 12740
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2810070344316922e-06,
      "loss": 1.2914,
      "step": 12741
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2795261014439098e-06,
      "loss": 1.1698,
      "step": 12742
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2780451684561273e-06,
      "loss": 1.5385,
      "step": 12743
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2765642354683452e-06,
      "loss": 1.0241,
      "step": 12744
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2750833024805627e-06,
      "loss": 1.2946,
      "step": 12745
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2736023694927805e-06,
      "loss": 1.3861,
      "step": 12746
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2721214365049984e-06,
      "loss": 1.1277,
      "step": 12747
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.270640503517216e-06,
      "loss": 1.1447,
      "step": 12748
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2691595705294339e-06,
      "loss": 1.2084,
      "step": 12749
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2676786375416514e-06,
      "loss": 1.3155,
      "step": 12750
    },
    {
      "epoch": 2.81,
      "eval_loss": NaN,
      "eval_runtime": 4.3796,
      "eval_samples_per_second": 352.548,
      "eval_steps_per_second": 44.068,
      "step": 12750
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2661977045538689e-06,
      "loss": 1.0747,
      "step": 12751
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2647167715660868e-06,
      "loss": 1.3158,
      "step": 12752
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2632358385783044e-06,
      "loss": 1.1303,
      "step": 12753
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.261754905590522e-06,
      "loss": 1.1076,
      "step": 12754
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.26027397260274e-06,
      "loss": 1.0215,
      "step": 12755
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2587930396149575e-06,
      "loss": 1.4773,
      "step": 12756
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.257312106627175e-06,
      "loss": 1.3038,
      "step": 12757
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.255831173639393e-06,
      "loss": 1.2178,
      "step": 12758
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2543502406516105e-06,
      "loss": 1.0859,
      "step": 12759
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2528693076638282e-06,
      "loss": 0.9538,
      "step": 12760
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.251388374676046e-06,
      "loss": 1.206,
      "step": 12761
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2499074416882637e-06,
      "loss": 1.2101,
      "step": 12762
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2484265087004814e-06,
      "loss": 1.3378,
      "step": 12763
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2469455757126992e-06,
      "loss": 1.2463,
      "step": 12764
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2454646427249167e-06,
      "loss": 1.2496,
      "step": 12765
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.2439837097371344e-06,
      "loss": 1.113,
      "step": 12766
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2425027767493521e-06,
      "loss": 1.3427,
      "step": 12767
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2410218437615699e-06,
      "loss": 1.2785,
      "step": 12768
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2395409107737876e-06,
      "loss": 0.9879,
      "step": 12769
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2380599777860053e-06,
      "loss": 1.1572,
      "step": 12770
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.236579044798223e-06,
      "loss": 1.0985,
      "step": 12771
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2350981118104406e-06,
      "loss": 1.313,
      "step": 12772
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2336171788226583e-06,
      "loss": 1.2621,
      "step": 12773
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.232136245834876e-06,
      "loss": 1.2943,
      "step": 12774
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2306553128470937e-06,
      "loss": 1.0327,
      "step": 12775
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2291743798593115e-06,
      "loss": 1.0153,
      "step": 12776
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2276934468715292e-06,
      "loss": 1.1512,
      "step": 12777
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.226212513883747e-06,
      "loss": 1.3912,
      "step": 12778
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2247315808959644e-06,
      "loss": 1.2612,
      "step": 12779
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2232506479081822e-06,
      "loss": 1.2414,
      "step": 12780
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2217697149204e-06,
      "loss": 1.0673,
      "step": 12781
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2202887819326176e-06,
      "loss": 1.3903,
      "step": 12782
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2188078489448354e-06,
      "loss": 1.5024,
      "step": 12783
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.217326915957053e-06,
      "loss": 1.0483,
      "step": 12784
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2158459829692708e-06,
      "loss": 0.9574,
      "step": 12785
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2143650499814883e-06,
      "loss": 1.1988,
      "step": 12786
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.212884116993706e-06,
      "loss": 1.2465,
      "step": 12787
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2114031840059238e-06,
      "loss": 1.1468,
      "step": 12788
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2099222510181415e-06,
      "loss": 1.0019,
      "step": 12789
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2084413180303593e-06,
      "loss": 1.3989,
      "step": 12790
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.206960385042577e-06,
      "loss": 1.1857,
      "step": 12791
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2054794520547947e-06,
      "loss": 1.2719,
      "step": 12792
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2039985190670122e-06,
      "loss": 1.1761,
      "step": 12793
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.20251758607923e-06,
      "loss": 1.0407,
      "step": 12794
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.2010366530914477e-06,
      "loss": 1.3218,
      "step": 12795
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1995557201036654e-06,
      "loss": 1.0436,
      "step": 12796
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1980747871158831e-06,
      "loss": 1.2359,
      "step": 12797
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1965938541281009e-06,
      "loss": 1.2187,
      "step": 12798
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1951129211403186e-06,
      "loss": 1.3389,
      "step": 12799
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1936319881525361e-06,
      "loss": 1.0322,
      "step": 12800
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1921510551647538e-06,
      "loss": 1.2903,
      "step": 12801
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1906701221769716e-06,
      "loss": 1.1162,
      "step": 12802
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1891891891891893e-06,
      "loss": 1.1094,
      "step": 12803
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.187708256201407e-06,
      "loss": 1.2085,
      "step": 12804
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1862273232136248e-06,
      "loss": 1.4563,
      "step": 12805
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1847463902258425e-06,
      "loss": 1.2427,
      "step": 12806
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.18326545723806e-06,
      "loss": 1.1909,
      "step": 12807
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1817845242502777e-06,
      "loss": 1.2623,
      "step": 12808
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1803035912624955e-06,
      "loss": 1.1708,
      "step": 12809
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.1788226582747132e-06,
      "loss": 1.3387,
      "step": 12810
    },
    {
      "epoch": 2.82,
      "learning_rate": 1.177341725286931e-06,
      "loss": 1.1197,
      "step": 12811
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1758607922991486e-06,
      "loss": 1.0633,
      "step": 12812
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1743798593113664e-06,
      "loss": 1.2076,
      "step": 12813
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1728989263235839e-06,
      "loss": 1.1897,
      "step": 12814
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1714179933358016e-06,
      "loss": 0.9792,
      "step": 12815
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1699370603480194e-06,
      "loss": 1.2818,
      "step": 12816
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.168456127360237e-06,
      "loss": 1.3912,
      "step": 12817
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1669751943724548e-06,
      "loss": 1.2349,
      "step": 12818
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1654942613846725e-06,
      "loss": 1.1438,
      "step": 12819
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1640133283968903e-06,
      "loss": 1.3062,
      "step": 12820
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1625323954091078e-06,
      "loss": 1.1736,
      "step": 12821
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1610514624213255e-06,
      "loss": 1.2098,
      "step": 12822
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1595705294335432e-06,
      "loss": 1.128,
      "step": 12823
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1580895964457608e-06,
      "loss": 1.0558,
      "step": 12824
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1566086634579787e-06,
      "loss": 1.3913,
      "step": 12825
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1551277304701964e-06,
      "loss": 1.1497,
      "step": 12826
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1536467974824142e-06,
      "loss": 1.0125,
      "step": 12827
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1521658644946317e-06,
      "loss": 1.0433,
      "step": 12828
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1506849315068494e-06,
      "loss": 1.1977,
      "step": 12829
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1492039985190671e-06,
      "loss": 1.1245,
      "step": 12830
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1477230655312846e-06,
      "loss": 1.5897,
      "step": 12831
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1462421325435024e-06,
      "loss": 1.2302,
      "step": 12832
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1447611995557203e-06,
      "loss": 1.1979,
      "step": 12833
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.143280266567938e-06,
      "loss": 1.2694,
      "step": 12834
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1417993335801556e-06,
      "loss": 1.1977,
      "step": 12835
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1403184005923733e-06,
      "loss": 1.3027,
      "step": 12836
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.138837467604591e-06,
      "loss": 1.2432,
      "step": 12837
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1373565346168085e-06,
      "loss": 1.3014,
      "step": 12838
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1358756016290263e-06,
      "loss": 1.1676,
      "step": 12839
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.134394668641244e-06,
      "loss": 1.1131,
      "step": 12840
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.132913735653462e-06,
      "loss": 0.9862,
      "step": 12841
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1314328026656794e-06,
      "loss": 1.3215,
      "step": 12842
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1299518696778972e-06,
      "loss": 1.3678,
      "step": 12843
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.128470936690115e-06,
      "loss": 0.9606,
      "step": 12844
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1269900037023324e-06,
      "loss": 1.3029,
      "step": 12845
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1255090707145502e-06,
      "loss": 1.1357,
      "step": 12846
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1240281377267679e-06,
      "loss": 1.2313,
      "step": 12847
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1225472047389856e-06,
      "loss": 1.2621,
      "step": 12848
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1210662717512033e-06,
      "loss": 1.2762,
      "step": 12849
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.119585338763421e-06,
      "loss": 1.1291,
      "step": 12850
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1181044057756388e-06,
      "loss": 1.1631,
      "step": 12851
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1166234727878563e-06,
      "loss": 1.1351,
      "step": 12852
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.115142539800074e-06,
      "loss": 1.2077,
      "step": 12853
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1136616068122918e-06,
      "loss": 1.2292,
      "step": 12854
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1121806738245095e-06,
      "loss": 1.1631,
      "step": 12855
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1106997408367272e-06,
      "loss": 1.1327,
      "step": 12856
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.109218807848945e-06,
      "loss": 1.1829,
      "step": 12857
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.1077378748611627e-06,
      "loss": 1.1676,
      "step": 12858
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.1062569418733802e-06,
      "loss": 1.2747,
      "step": 12859
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.104776008885598e-06,
      "loss": 1.2815,
      "step": 12860
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.1032950758978157e-06,
      "loss": 1.1791,
      "step": 12861
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.1018141429100334e-06,
      "loss": 1.2493,
      "step": 12862
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.1003332099222511e-06,
      "loss": 1.1028,
      "step": 12863
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0988522769344688e-06,
      "loss": 1.3482,
      "step": 12864
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0973713439466866e-06,
      "loss": 1.0106,
      "step": 12865
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.095890410958904e-06,
      "loss": 1.1084,
      "step": 12866
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0944094779711218e-06,
      "loss": 1.1183,
      "step": 12867
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0929285449833395e-06,
      "loss": 1.0319,
      "step": 12868
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0914476119955573e-06,
      "loss": 1.1605,
      "step": 12869
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.089966679007775e-06,
      "loss": 1.1279,
      "step": 12870
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0884857460199927e-06,
      "loss": 1.2792,
      "step": 12871
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0870048130322105e-06,
      "loss": 1.1613,
      "step": 12872
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.085523880044428e-06,
      "loss": 1.2894,
      "step": 12873
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0840429470566457e-06,
      "loss": 1.3137,
      "step": 12874
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0825620140688634e-06,
      "loss": 1.4843,
      "step": 12875
    },
    {
      "epoch": 2.84,
      "eval_loss": NaN,
      "eval_runtime": 4.3777,
      "eval_samples_per_second": 352.693,
      "eval_steps_per_second": 44.087,
      "step": 12875
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0810810810810812e-06,
      "loss": 1.2426,
      "step": 12876
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0796001480932989e-06,
      "loss": 1.4277,
      "step": 12877
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0781192151055166e-06,
      "loss": 1.1452,
      "step": 12878
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0766382821177343e-06,
      "loss": 1.0248,
      "step": 12879
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0751573491299519e-06,
      "loss": 1.1481,
      "step": 12880
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0736764161421696e-06,
      "loss": 1.1297,
      "step": 12881
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0721954831543873e-06,
      "loss": 1.1497,
      "step": 12882
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.070714550166605e-06,
      "loss": 1.1423,
      "step": 12883
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0692336171788228e-06,
      "loss": 1.3982,
      "step": 12884
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0677526841910405e-06,
      "loss": 1.163,
      "step": 12885
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0662717512032582e-06,
      "loss": 1.3971,
      "step": 12886
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0647908182154758e-06,
      "loss": 1.4777,
      "step": 12887
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0633098852276935e-06,
      "loss": 1.1986,
      "step": 12888
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0618289522399112e-06,
      "loss": 1.1649,
      "step": 12889
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.060348019252129e-06,
      "loss": 1.2381,
      "step": 12890
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0588670862643467e-06,
      "loss": 1.1375,
      "step": 12891
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0573861532765644e-06,
      "loss": 1.5125,
      "step": 12892
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0559052202887821e-06,
      "loss": 1.1695,
      "step": 12893
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0544242873009996e-06,
      "loss": 1.3592,
      "step": 12894
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0529433543132174e-06,
      "loss": 0.9778,
      "step": 12895
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.051462421325435e-06,
      "loss": 1.1938,
      "step": 12896
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0499814883376528e-06,
      "loss": 1.2154,
      "step": 12897
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0485005553498706e-06,
      "loss": 1.2793,
      "step": 12898
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0470196223620883e-06,
      "loss": 1.2525,
      "step": 12899
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.045538689374306e-06,
      "loss": 1.0702,
      "step": 12900
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0440577563865235e-06,
      "loss": 1.4761,
      "step": 12901
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.0425768233987413e-06,
      "loss": 1.3206,
      "step": 12902
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.041095890410959e-06,
      "loss": 1.1112,
      "step": 12903
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0396149574231767e-06,
      "loss": 1.2208,
      "step": 12904
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0381340244353944e-06,
      "loss": 1.2317,
      "step": 12905
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0366530914476122e-06,
      "loss": 1.1252,
      "step": 12906
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.03517215845983e-06,
      "loss": 1.0681,
      "step": 12907
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0336912254720474e-06,
      "loss": 1.0994,
      "step": 12908
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0322102924842652e-06,
      "loss": 1.1965,
      "step": 12909
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0307293594964829e-06,
      "loss": 0.9913,
      "step": 12910
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0292484265087006e-06,
      "loss": 1.4105,
      "step": 12911
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0277674935209181e-06,
      "loss": 1.0645,
      "step": 12912
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.026286560533136e-06,
      "loss": 0.8622,
      "step": 12913
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0248056275453538e-06,
      "loss": 1.2002,
      "step": 12914
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0233246945575713e-06,
      "loss": 1.3467,
      "step": 12915
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.021843761569789e-06,
      "loss": 1.065,
      "step": 12916
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0203628285820068e-06,
      "loss": 1.1776,
      "step": 12917
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0188818955942245e-06,
      "loss": 1.29,
      "step": 12918
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.017400962606442e-06,
      "loss": 1.1914,
      "step": 12919
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0159200296186597e-06,
      "loss": 1.2211,
      "step": 12920
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0144390966308777e-06,
      "loss": 1.4088,
      "step": 12921
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0129581636430952e-06,
      "loss": 1.1567,
      "step": 12922
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.011477230655313e-06,
      "loss": 1.1913,
      "step": 12923
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0099962976675307e-06,
      "loss": 1.2114,
      "step": 12924
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0085153646797484e-06,
      "loss": 1.3053,
      "step": 12925
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.007034431691966e-06,
      "loss": 0.9915,
      "step": 12926
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0055534987041836e-06,
      "loss": 0.9713,
      "step": 12927
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0040725657164014e-06,
      "loss": 1.1407,
      "step": 12928
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.002591632728619e-06,
      "loss": 1.2409,
      "step": 12929
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.0011106997408368e-06,
      "loss": 1.2883,
      "step": 12930
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.996297667530545e-07,
      "loss": 1.2752,
      "step": 12931
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.98148833765272e-07,
      "loss": 1.0142,
      "step": 12932
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.966679007774898e-07,
      "loss": 1.2371,
      "step": 12933
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.951869677897075e-07,
      "loss": 1.0393,
      "step": 12934
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.937060348019252e-07,
      "loss": 1.3066,
      "step": 12935
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.92225101814143e-07,
      "loss": 1.0898,
      "step": 12936
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.907441688263607e-07,
      "loss": 1.1402,
      "step": 12937
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.892632358385784e-07,
      "loss": 1.1694,
      "step": 12938
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.87782302850796e-07,
      "loss": 1.1886,
      "step": 12939
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.863013698630137e-07,
      "loss": 1.2247,
      "step": 12940
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.848204368752314e-07,
      "loss": 1.3813,
      "step": 12941
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.833395038874491e-07,
      "loss": 1.3811,
      "step": 12942
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.818585708996669e-07,
      "loss": 1.3061,
      "step": 12943
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.803776379118846e-07,
      "loss": 1.2347,
      "step": 12944
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.788967049241023e-07,
      "loss": 1.3527,
      "step": 12945
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.774157719363198e-07,
      "loss": 1.1822,
      "step": 12946
    },
    {
      "epoch": 2.85,
      "learning_rate": 9.759348389485376e-07,
      "loss": 1.2055,
      "step": 12947
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.744539059607553e-07,
      "loss": 1.199,
      "step": 12948
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.72972972972973e-07,
      "loss": 1.2807,
      "step": 12949
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.714920399851908e-07,
      "loss": 1.1865,
      "step": 12950
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.700111069974085e-07,
      "loss": 1.2774,
      "step": 12951
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.685301740096262e-07,
      "loss": 1.2402,
      "step": 12952
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.670492410218437e-07,
      "loss": 1.1456,
      "step": 12953
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.655683080340615e-07,
      "loss": 1.1888,
      "step": 12954
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.640873750462792e-07,
      "loss": 1.1571,
      "step": 12955
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.62606442058497e-07,
      "loss": 1.4676,
      "step": 12956
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.611255090707146e-07,
      "loss": 1.1036,
      "step": 12957
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.596445760829324e-07,
      "loss": 1.195,
      "step": 12958
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.5816364309515e-07,
      "loss": 1.3679,
      "step": 12959
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.566827101073676e-07,
      "loss": 1.1084,
      "step": 12960
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.552017771195853e-07,
      "loss": 1.2975,
      "step": 12961
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.53720844131803e-07,
      "loss": 1.1633,
      "step": 12962
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.522399111440209e-07,
      "loss": 1.1501,
      "step": 12963
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.507589781562384e-07,
      "loss": 1.2482,
      "step": 12964
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.492780451684562e-07,
      "loss": 1.1591,
      "step": 12965
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.47797112180674e-07,
      "loss": 1.5288,
      "step": 12966
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.463161791928915e-07,
      "loss": 1.3682,
      "step": 12967
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.448352462051092e-07,
      "loss": 1.3895,
      "step": 12968
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.43354313217327e-07,
      "loss": 1.3641,
      "step": 12969
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.418733802295448e-07,
      "loss": 1.1797,
      "step": 12970
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.403924472417623e-07,
      "loss": 1.2815,
      "step": 12971
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.3891151425398e-07,
      "loss": 1.0924,
      "step": 12972
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.374305812661978e-07,
      "loss": 1.3069,
      "step": 12973
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.359496482784154e-07,
      "loss": 1.3697,
      "step": 12974
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.344687152906331e-07,
      "loss": 1.2354,
      "step": 12975
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.329877823028509e-07,
      "loss": 1.2209,
      "step": 12976
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.315068493150686e-07,
      "loss": 0.9845,
      "step": 12977
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.300259163272862e-07,
      "loss": 1.1789,
      "step": 12978
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.285449833395039e-07,
      "loss": 1.2153,
      "step": 12979
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.270640503517217e-07,
      "loss": 1.1712,
      "step": 12980
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.255831173639393e-07,
      "loss": 1.0457,
      "step": 12981
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.24102184376157e-07,
      "loss": 1.2779,
      "step": 12982
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.226212513883747e-07,
      "loss": 1.2256,
      "step": 12983
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.211403184005925e-07,
      "loss": 1.3714,
      "step": 12984
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.196593854128101e-07,
      "loss": 1.2465,
      "step": 12985
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.181784524250278e-07,
      "loss": 1.1936,
      "step": 12986
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.166975194372456e-07,
      "loss": 1.1073,
      "step": 12987
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.152165864494632e-07,
      "loss": 1.0436,
      "step": 12988
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.137356534616809e-07,
      "loss": 1.0065,
      "step": 12989
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.122547204738986e-07,
      "loss": 1.2758,
      "step": 12990
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.107737874861164e-07,
      "loss": 1.1219,
      "step": 12991
    },
    {
      "epoch": 2.86,
      "learning_rate": 9.09292854498334e-07,
      "loss": 1.2045,
      "step": 12992
    },
    {
      "epoch": 2.87,
      "learning_rate": 9.078119215105517e-07,
      "loss": 0.9958,
      "step": 12993
    },
    {
      "epoch": 2.87,
      "learning_rate": 9.063309885227694e-07,
      "loss": 1.0317,
      "step": 12994
    },
    {
      "epoch": 2.87,
      "learning_rate": 9.048500555349871e-07,
      "loss": 1.2522,
      "step": 12995
    },
    {
      "epoch": 2.87,
      "learning_rate": 9.033691225472048e-07,
      "loss": 1.4168,
      "step": 12996
    },
    {
      "epoch": 2.87,
      "learning_rate": 9.018881895594225e-07,
      "loss": 0.995,
      "step": 12997
    },
    {
      "epoch": 2.87,
      "learning_rate": 9.004072565716402e-07,
      "loss": 1.2216,
      "step": 12998
    },
    {
      "epoch": 2.87,
      "learning_rate": 8.989263235838579e-07,
      "loss": 1.1385,
      "step": 12999
    },
    {
      "epoch": 2.87,
      "learning_rate": 8.974453905960756e-07,
      "loss": 1.2742,
      "step": 13000
    },
    {
      "epoch": 2.87,
      "eval_loss": NaN,
      "eval_runtime": 4.3713,
      "eval_samples_per_second": 353.215,
      "eval_steps_per_second": 44.152,
      "step": 13000
    }
  ],
  "logging_steps": 1,
  "max_steps": 13605,
  "num_train_epochs": 3,
  "save_steps": 250,
  "total_flos": 1.465366615439704e+18,
  "trial_name": null,
  "trial_params": null
}
