{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 785,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032,
      "grad_norm": 5.24710145178152,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.7913,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.21088887751102448,
      "step": 5,
      "valid_targets_mean": 5874.7,
      "valid_targets_min": 2001
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.3277670536622015,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.7652,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.18997091054916382,
      "step": 10,
      "valid_targets_mean": 6202.2,
      "valid_targets_min": 3824
    },
    {
      "epoch": 0.096,
      "grad_norm": 2.0433898822364345,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.7275,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.19564616680145264,
      "step": 15,
      "valid_targets_mean": 5657.1,
      "valid_targets_min": 3086
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.8872433295444813,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.6919,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17012247443199158,
      "step": 20,
      "valid_targets_mean": 5604.0,
      "valid_targets_min": 1367
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7934069909160805,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.6754,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17535243928432465,
      "step": 25,
      "valid_targets_mean": 5755.4,
      "valid_targets_min": 2008
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.5185660654502818,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.6463,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16546516120433807,
      "step": 30,
      "valid_targets_mean": 5578.9,
      "valid_targets_min": 2239
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.4442582062801901,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.6187,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14229542016983032,
      "step": 35,
      "valid_targets_mean": 5112.4,
      "valid_targets_min": 1316
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.34929437826789717,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.6019,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15741989016532898,
      "step": 40,
      "valid_targets_mean": 5765.8,
      "valid_targets_min": 1255
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.32295231079081044,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.5938,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14423394203186035,
      "step": 45,
      "valid_targets_mean": 5555.3,
      "valid_targets_min": 3489
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.2881101734804579,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.5733,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1374053657054901,
      "step": 50,
      "valid_targets_mean": 5621.2,
      "valid_targets_min": 2075
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.28148191610821155,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.5647,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11957500874996185,
      "step": 55,
      "valid_targets_mean": 4791.2,
      "valid_targets_min": 1195
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.2765176294470177,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.5653,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13385950028896332,
      "step": 60,
      "valid_targets_mean": 5334.9,
      "valid_targets_min": 860
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.2659158971926803,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.5397,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13042165338993073,
      "step": 65,
      "valid_targets_mean": 5692.4,
      "valid_targets_min": 1083
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.27296487641491296,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.541,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13398760557174683,
      "step": 70,
      "valid_targets_mean": 5619.6,
      "valid_targets_min": 2385
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.25557447966339214,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.5288,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1376691460609436,
      "step": 75,
      "valid_targets_mean": 5343.9,
      "valid_targets_min": 1362
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.2676449550715318,
      "learning_rate": 4e-05,
      "loss": 0.529,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1336148977279663,
      "step": 80,
      "valid_targets_mean": 5567.1,
      "valid_targets_min": 2955
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.26900294190979,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.5322,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13481895625591278,
      "step": 85,
      "valid_targets_mean": 5472.6,
      "valid_targets_min": 3818
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.27776130098530794,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.5039,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12478074431419373,
      "step": 90,
      "valid_targets_mean": 5482.8,
      "valid_targets_min": 2955
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.2713930326036343,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.5109,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12081289291381836,
      "step": 95,
      "valid_targets_mean": 5249.3,
      "valid_targets_min": 1688
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.27752374907774013,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.5025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1324721723794937,
      "step": 100,
      "valid_targets_mean": 5962.4,
      "valid_targets_min": 2748
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.267499936413584,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.5107,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13119545578956604,
      "step": 105,
      "valid_targets_mean": 5409.4,
      "valid_targets_min": 435
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.23766079629521358,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.5029,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12771186232566833,
      "step": 110,
      "valid_targets_mean": 5800.8,
      "valid_targets_min": 1929
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.25772697025474894,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.5121,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12370181828737259,
      "step": 115,
      "valid_targets_mean": 5479.6,
      "valid_targets_min": 1022
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.26022439252733154,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.4953,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12205180525779724,
      "step": 120,
      "valid_targets_mean": 5375.6,
      "valid_targets_min": 2871
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.28955279552081553,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.5033,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13474714756011963,
      "step": 125,
      "valid_targets_mean": 5777.4,
      "valid_targets_min": 4832
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.2809383519265258,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.4856,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11271673440933228,
      "step": 130,
      "valid_targets_mean": 5723.9,
      "valid_targets_min": 3754
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.24936027543074415,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.4868,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12169145047664642,
      "step": 135,
      "valid_targets_mean": 5967.4,
      "valid_targets_min": 3602
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.2717766834302193,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.4912,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13354289531707764,
      "step": 140,
      "valid_targets_mean": 5569.7,
      "valid_targets_min": 1666
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.3261900724811267,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.4942,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1227942705154419,
      "step": 145,
      "valid_targets_mean": 5438.8,
      "valid_targets_min": 3413
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.26705134476493997,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.485,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12847238779067993,
      "step": 150,
      "valid_targets_mean": 5774.5,
      "valid_targets_min": 3118
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.2675570325178453,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.4827,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12189042568206787,
      "step": 155,
      "valid_targets_mean": 6038.6,
      "valid_targets_min": 2582
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.2711974624490461,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.4836,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11894658207893372,
      "step": 160,
      "valid_targets_mean": 6018.4,
      "valid_targets_min": 4144
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.2511797202880708,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.4866,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12989544868469238,
      "step": 165,
      "valid_targets_mean": 6497.6,
      "valid_targets_min": 4294
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.3220210091642244,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.4743,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1322012096643448,
      "step": 170,
      "valid_targets_mean": 5688.2,
      "valid_targets_min": 4045
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.2829880452875022,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.481,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13040798902511597,
      "step": 175,
      "valid_targets_mean": 5453.4,
      "valid_targets_min": 3778
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.27722869676516065,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.4696,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1015506237745285,
      "step": 180,
      "valid_targets_mean": 5323.6,
      "valid_targets_min": 1515
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.3120730587917303,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.4758,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12049368768930435,
      "step": 185,
      "valid_targets_mean": 5981.9,
      "valid_targets_min": 3729
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.3177904238417455,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.4805,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11169823259115219,
      "step": 190,
      "valid_targets_mean": 5015.9,
      "valid_targets_min": 1686
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.2820969266968138,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.4711,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1114407479763031,
      "step": 195,
      "valid_targets_mean": 4919.9,
      "valid_targets_min": 465
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.29829183019272854,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.4838,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13220760226249695,
      "step": 200,
      "valid_targets_mean": 5712.1,
      "valid_targets_min": 2519
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.3324584176695949,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.4693,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1180618479847908,
      "step": 205,
      "valid_targets_mean": 5392.0,
      "valid_targets_min": 3327
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.27642608588216105,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.4784,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12334628403186798,
      "step": 210,
      "valid_targets_mean": 6032.2,
      "valid_targets_min": 2884
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.3031104214893501,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.4862,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12392963469028473,
      "step": 215,
      "valid_targets_mean": 5173.2,
      "valid_targets_min": 2773
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.2627321589062865,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.4717,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10038068890571594,
      "step": 220,
      "valid_targets_mean": 4871.1,
      "valid_targets_min": 929
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.276565357842663,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.477,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12069591879844666,
      "step": 225,
      "valid_targets_mean": 6686.1,
      "valid_targets_min": 5410
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.26436389532991533,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.4737,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11359324306249619,
      "step": 230,
      "valid_targets_mean": 5082.8,
      "valid_targets_min": 3308
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.2796465196510237,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.4631,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11204469203948975,
      "step": 235,
      "valid_targets_mean": 6092.7,
      "valid_targets_min": 693
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.3309030224476089,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.4686,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11983086913824081,
      "step": 240,
      "valid_targets_mean": 6208.1,
      "valid_targets_min": 2743
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.2713704011247261,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.4724,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11787189543247223,
      "step": 245,
      "valid_targets_mean": 5622.1,
      "valid_targets_min": 3834
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.25249294559973856,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.4682,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11265088617801666,
      "step": 250,
      "valid_targets_mean": 5751.2,
      "valid_targets_min": 3118
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.2952259016724535,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.4679,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11052439361810684,
      "step": 255,
      "valid_targets_mean": 5447.6,
      "valid_targets_min": 2671
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.2589109654816995,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.4686,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10572341084480286,
      "step": 260,
      "valid_targets_mean": 5510.8,
      "valid_targets_min": 3259
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.2644845788435991,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.4638,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11645673960447311,
      "step": 265,
      "valid_targets_mean": 5364.7,
      "valid_targets_min": 1097
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.2551481979485969,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.4628,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11428695172071457,
      "step": 270,
      "valid_targets_mean": 5801.9,
      "valid_targets_min": 2008
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.2781800499632048,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.4727,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12068630754947662,
      "step": 275,
      "valid_targets_mean": 5926.0,
      "valid_targets_min": 3063
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.24626783491400464,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.4623,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11117767542600632,
      "step": 280,
      "valid_targets_mean": 5921.8,
      "valid_targets_min": 2960
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.2760365203620935,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.4616,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11723286658525467,
      "step": 285,
      "valid_targets_mean": 5935.3,
      "valid_targets_min": 3083
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.27282562791523385,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.466,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10799738764762878,
      "step": 290,
      "valid_targets_mean": 4728.4,
      "valid_targets_min": 2039
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.270151566840584,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.4676,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12066589295864105,
      "step": 295,
      "valid_targets_mean": 5457.9,
      "valid_targets_min": 2132
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.270240069931949,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.4607,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12481565773487091,
      "step": 300,
      "valid_targets_mean": 6084.4,
      "valid_targets_min": 3841
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.2607976462751014,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.456,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10547001659870148,
      "step": 305,
      "valid_targets_mean": 5309.6,
      "valid_targets_min": 1982
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.2801317764093328,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.4609,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11671700328588486,
      "step": 310,
      "valid_targets_mean": 5753.4,
      "valid_targets_min": 2814
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.2778483897221595,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.4595,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12135300040245056,
      "step": 315,
      "valid_targets_mean": 5642.1,
      "valid_targets_min": 1524
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.27820443433191205,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.4492,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11907652765512466,
      "step": 320,
      "valid_targets_mean": 5979.9,
      "valid_targets_min": 3237
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.27108436735341157,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.4551,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11400112509727478,
      "step": 325,
      "valid_targets_mean": 5907.1,
      "valid_targets_min": 1687
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.27844649674754907,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.4487,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.113765649497509,
      "step": 330,
      "valid_targets_mean": 5919.2,
      "valid_targets_min": 3819
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.2744264229880744,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.4505,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11026716232299805,
      "step": 335,
      "valid_targets_mean": 4980.7,
      "valid_targets_min": 3088
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.2856089911494548,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.4538,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12398375570774078,
      "step": 340,
      "valid_targets_mean": 5893.3,
      "valid_targets_min": 1951
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.2866894573358364,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.4563,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12197114527225494,
      "step": 345,
      "valid_targets_mean": 5729.6,
      "valid_targets_min": 1334
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.2573254590182236,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.4497,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1142326220870018,
      "step": 350,
      "valid_targets_mean": 6702.6,
      "valid_targets_min": 3271
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.29771620142160105,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.4541,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11867627501487732,
      "step": 355,
      "valid_targets_mean": 5739.6,
      "valid_targets_min": 1518
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.261753688103989,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.452,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1085813120007515,
      "step": 360,
      "valid_targets_mean": 5226.2,
      "valid_targets_min": 2194
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.25894325787280104,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.4574,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11135275661945343,
      "step": 365,
      "valid_targets_mean": 5223.6,
      "valid_targets_min": 271
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.2412639805817092,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.453,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11326012015342712,
      "step": 370,
      "valid_targets_mean": 5809.9,
      "valid_targets_min": 3821
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.27318877008295256,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.4444,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12262687087059021,
      "step": 375,
      "valid_targets_mean": 6501.4,
      "valid_targets_min": 3667
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.26904838926012503,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.4481,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11483996361494064,
      "step": 380,
      "valid_targets_mean": 5609.5,
      "valid_targets_min": 3066
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.2665045221550359,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.4594,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11784804612398148,
      "step": 385,
      "valid_targets_mean": 5498.8,
      "valid_targets_min": 2572
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.252522725335052,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.4542,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11689453572034836,
      "step": 390,
      "valid_targets_mean": 5693.0,
      "valid_targets_min": 2832
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.2507034098685498,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.448,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11683270335197449,
      "step": 395,
      "valid_targets_mean": 5887.5,
      "valid_targets_min": 3706
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.23529071762965234,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.4442,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10294480621814728,
      "step": 400,
      "valid_targets_mean": 5279.1,
      "valid_targets_min": 1143
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.27536147692711077,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.4532,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12593109905719757,
      "step": 405,
      "valid_targets_mean": 6261.8,
      "valid_targets_min": 4713
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.2592631543468272,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.4504,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10663522779941559,
      "step": 410,
      "valid_targets_mean": 5489.7,
      "valid_targets_min": 1218
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.26112482485360655,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.4527,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11844196170568466,
      "step": 415,
      "valid_targets_mean": 6735.8,
      "valid_targets_min": 3925
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.26689519246895416,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.4518,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12620952725410461,
      "step": 420,
      "valid_targets_mean": 5928.7,
      "valid_targets_min": 3496
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.25760408233245335,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.4555,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08366312831640244,
      "step": 425,
      "valid_targets_mean": 3737.7,
      "valid_targets_min": 738
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.25686185255102495,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.4514,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1142929419875145,
      "step": 430,
      "valid_targets_mean": 5334.4,
      "valid_targets_min": 1515
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.27245429840074986,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.4464,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1127873957157135,
      "step": 435,
      "valid_targets_mean": 6076.6,
      "valid_targets_min": 1780
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.2427804170228958,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.449,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12002001702785492,
      "step": 440,
      "valid_targets_mean": 6298.9,
      "valid_targets_min": 4125
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.2670724501923031,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.45,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11933128535747528,
      "step": 445,
      "valid_targets_mean": 5839.9,
      "valid_targets_min": 2675
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.24592387963036164,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.4551,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10749687254428864,
      "step": 450,
      "valid_targets_mean": 5285.9,
      "valid_targets_min": 2874
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.2561803177618425,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.4493,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11277574300765991,
      "step": 455,
      "valid_targets_mean": 5292.3,
      "valid_targets_min": 2180
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.27119960756548817,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.4504,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12340002506971359,
      "step": 460,
      "valid_targets_mean": 5302.8,
      "valid_targets_min": 846
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.24611436075482088,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.4511,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1188078373670578,
      "step": 465,
      "valid_targets_mean": 6020.6,
      "valid_targets_min": 3642
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.26014356778475395,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.4483,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.122434101998806,
      "step": 470,
      "valid_targets_mean": 5858.0,
      "valid_targets_min": 3103
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.25505384170050344,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.4449,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11307892203330994,
      "step": 475,
      "valid_targets_mean": 5812.4,
      "valid_targets_min": 1881
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.2526560000616734,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.4394,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10069246590137482,
      "step": 480,
      "valid_targets_mean": 5966.9,
      "valid_targets_min": 1574
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.25178932184474906,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.447,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.112733393907547,
      "step": 485,
      "valid_targets_mean": 5253.1,
      "valid_targets_min": 1674
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.24800375473384853,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.4441,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10786256194114685,
      "step": 490,
      "valid_targets_mean": 5875.4,
      "valid_targets_min": 4090
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.24892569758159228,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.4353,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10156938433647156,
      "step": 495,
      "valid_targets_mean": 4843.1,
      "valid_targets_min": 1316
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.25477312307497063,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.4427,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10294046252965927,
      "step": 500,
      "valid_targets_mean": 5156.9,
      "valid_targets_min": 915
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.2438583455977595,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.4341,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11227782070636749,
      "step": 505,
      "valid_targets_mean": 5543.9,
      "valid_targets_min": 2788
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.2432731418382833,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.4368,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11503825336694717,
      "step": 510,
      "valid_targets_mean": 6555.9,
      "valid_targets_min": 3981
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.25520474007057975,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.4361,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10623282194137573,
      "step": 515,
      "valid_targets_mean": 5709.8,
      "valid_targets_min": 2300
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.23518767138484464,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.4383,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11089183390140533,
      "step": 520,
      "valid_targets_mean": 5958.1,
      "valid_targets_min": 3118
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.2647501245074642,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.4468,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1223001778125763,
      "step": 525,
      "valid_targets_mean": 5351.3,
      "valid_targets_min": 1465
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.2573457479976076,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.4428,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1174464300274849,
      "step": 530,
      "valid_targets_mean": 5884.6,
      "valid_targets_min": 3402
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.23854531691277037,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.4386,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10301633179187775,
      "step": 535,
      "valid_targets_mean": 5470.4,
      "valid_targets_min": 368
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.23204549273866737,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.4326,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10386782139539719,
      "step": 540,
      "valid_targets_mean": 5608.4,
      "valid_targets_min": 1516
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.24254831311806727,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.4406,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11488397419452667,
      "step": 545,
      "valid_targets_mean": 5481.1,
      "valid_targets_min": 3655
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.23160404922336977,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.4496,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11126826703548431,
      "step": 550,
      "valid_targets_mean": 5314.2,
      "valid_targets_min": 1444
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.24735196100744516,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.4332,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.102394238114357,
      "step": 555,
      "valid_targets_mean": 4585.0,
      "valid_targets_min": 870
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.24107940600744615,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11586499214172363,
      "step": 560,
      "valid_targets_mean": 5674.2,
      "valid_targets_min": 1575
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.24261915049213512,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.4402,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1045859158039093,
      "step": 565,
      "valid_targets_mean": 5670.9,
      "valid_targets_min": 2223
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.24933380090093396,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.4336,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09782751649618149,
      "step": 570,
      "valid_targets_mean": 5943.5,
      "valid_targets_min": 2789
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.24876871762840735,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.4443,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1115277111530304,
      "step": 575,
      "valid_targets_mean": 5512.4,
      "valid_targets_min": 2924
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.26310062940650597,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.4436,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10961464047431946,
      "step": 580,
      "valid_targets_mean": 5330.4,
      "valid_targets_min": 2085
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.25870239473978524,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.441,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11257312446832657,
      "step": 585,
      "valid_targets_mean": 5724.1,
      "valid_targets_min": 3871
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.23326000227040175,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.4449,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1034560352563858,
      "step": 590,
      "valid_targets_mean": 5530.9,
      "valid_targets_min": 2557
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.23282685187336052,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.4435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10593272745609283,
      "step": 595,
      "valid_targets_mean": 5772.1,
      "valid_targets_min": 1581
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.22765313286260372,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.4498,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10898633301258087,
      "step": 600,
      "valid_targets_mean": 6027.4,
      "valid_targets_min": 4510
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.2655078342405895,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.4476,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10464062541723251,
      "step": 605,
      "valid_targets_mean": 5047.8,
      "valid_targets_min": 2725
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.23245435258699704,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.4355,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10366545617580414,
      "step": 610,
      "valid_targets_mean": 5696.2,
      "valid_targets_min": 1771
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.259870240607499,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.4477,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11879684031009674,
      "step": 615,
      "valid_targets_mean": 5330.5,
      "valid_targets_min": 719
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.24926514499055483,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.4407,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11707712709903717,
      "step": 620,
      "valid_targets_mean": 5702.8,
      "valid_targets_min": 3274
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.24621203257822732,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.4457,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10299795866012573,
      "step": 625,
      "valid_targets_mean": 5261.9,
      "valid_targets_min": 2518
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.29032992924768214,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.4422,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12003006041049957,
      "step": 630,
      "valid_targets_mean": 5993.6,
      "valid_targets_min": 3298
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.23011309001967206,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.443,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12141761928796768,
      "step": 635,
      "valid_targets_mean": 6614.4,
      "valid_targets_min": 3114
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.23010955969434194,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.4376,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11637435853481293,
      "step": 640,
      "valid_targets_mean": 5707.2,
      "valid_targets_min": 4154
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.22749349183887843,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.4384,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09296316653490067,
      "step": 645,
      "valid_targets_mean": 5367.4,
      "valid_targets_min": 523
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.23494376876007655,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.4377,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10853186249732971,
      "step": 650,
      "valid_targets_mean": 5471.6,
      "valid_targets_min": 1855
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.2273668884710314,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.4321,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11260022968053818,
      "step": 655,
      "valid_targets_mean": 5910.9,
      "valid_targets_min": 797
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.25666942843106755,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.4417,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11119656264781952,
      "step": 660,
      "valid_targets_mean": 5067.0,
      "valid_targets_min": 2851
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.23427420038374935,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.4265,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10912980884313583,
      "step": 665,
      "valid_targets_mean": 6025.1,
      "valid_targets_min": 3327
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.2501556392462366,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.4332,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.120496466755867,
      "step": 670,
      "valid_targets_mean": 5586.7,
      "valid_targets_min": 1622
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.23266280463889957,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.4374,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10693095624446869,
      "step": 675,
      "valid_targets_mean": 5516.1,
      "valid_targets_min": 707
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.25949014222045463,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.4437,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10792089253664017,
      "step": 680,
      "valid_targets_mean": 5310.1,
      "valid_targets_min": 915
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.23465382879991956,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.4337,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11585308611392975,
      "step": 685,
      "valid_targets_mean": 5791.5,
      "valid_targets_min": 4122
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.2215726331342345,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.4343,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10746041685342789,
      "step": 690,
      "valid_targets_mean": 5557.8,
      "valid_targets_min": 1362
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.22743877753089006,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.4329,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09633181244134903,
      "step": 695,
      "valid_targets_mean": 5391.2,
      "valid_targets_min": 1097
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.23932325312058522,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.4359,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11483240872621536,
      "step": 700,
      "valid_targets_mean": 5383.9,
      "valid_targets_min": 4166
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.22913767958555759,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.4378,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10419732332229614,
      "step": 705,
      "valid_targets_mean": 5465.1,
      "valid_targets_min": 1829
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.24074376103641706,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.4422,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10467822849750519,
      "step": 710,
      "valid_targets_mean": 5554.9,
      "valid_targets_min": 2465
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.2285117185268389,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.4382,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09651068598031998,
      "step": 715,
      "valid_targets_mean": 4730.2,
      "valid_targets_min": 655
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.23306160374824716,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.4342,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09519432485103607,
      "step": 720,
      "valid_targets_mean": 5076.6,
      "valid_targets_min": 446
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.2238739279040776,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.4402,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10766564309597015,
      "step": 725,
      "valid_targets_mean": 5361.7,
      "valid_targets_min": 2689
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.22340741915443632,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.438,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10900825262069702,
      "step": 730,
      "valid_targets_mean": 5918.2,
      "valid_targets_min": 3920
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.2420478067191904,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.4273,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11152715235948563,
      "step": 735,
      "valid_targets_mean": 6090.6,
      "valid_targets_min": 2899
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.22879981625600163,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.4321,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11784280836582184,
      "step": 740,
      "valid_targets_mean": 5993.5,
      "valid_targets_min": 1574
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.21838835580462193,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.4324,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10759489238262177,
      "step": 745,
      "valid_targets_mean": 5427.1,
      "valid_targets_min": 1951
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.23659034192495684,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.4377,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12015394866466522,
      "step": 750,
      "valid_targets_mean": 6135.6,
      "valid_targets_min": 3086
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.24249273716948083,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.434,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11292092502117157,
      "step": 755,
      "valid_targets_mean": 5941.0,
      "valid_targets_min": 738
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.2332679048833938,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.4337,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11557448655366898,
      "step": 760,
      "valid_targets_mean": 5841.6,
      "valid_targets_min": 3854
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.22456272081842824,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.4439,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09964489936828613,
      "step": 765,
      "valid_targets_mean": 5347.1,
      "valid_targets_min": 3272
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.2316304318267485,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.4357,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09679272025823593,
      "step": 770,
      "valid_targets_mean": 4909.6,
      "valid_targets_min": 667
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.24136995238857187,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.4435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10119546204805374,
      "step": 775,
      "valid_targets_mean": 4710.8,
      "valid_targets_min": 970
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.24723332555052074,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.438,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11632305383682251,
      "step": 780,
      "valid_targets_mean": 4985.4,
      "valid_targets_min": 368
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.44517301236898366,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.4408,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4259156584739685,
      "step": 785,
      "valid_targets_mean": 6322.3,
      "valid_targets_min": 1465
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4259156584739685,
      "step": 785,
      "total_flos": 1.9782193358307328e+18,
      "train_loss": 0.4723992296085236,
      "train_runtime": 18119.4294,
      "train_samples_per_second": 2.759,
      "train_steps_per_second": 0.043,
      "valid_targets_mean": 6322.3,
      "valid_targets_min": 1465
    }
  ],
  "logging_steps": 5,
  "max_steps": 785,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9782193358307328e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}