| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 820, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.012195121951219513, | |
| "grad_norm": 460.0, | |
| "learning_rate": 2.4390243902439027e-06, | |
| "loss": 48.1093, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06097560975609756, | |
| "grad_norm": 360.0, | |
| "learning_rate": 1.2195121951219513e-05, | |
| "loss": 42.5766, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12195121951219512, | |
| "grad_norm": 187.0, | |
| "learning_rate": 2.4390243902439026e-05, | |
| "loss": 35.1457, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.18292682926829268, | |
| "grad_norm": 31.375, | |
| "learning_rate": 3.6585365853658535e-05, | |
| "loss": 21.8793, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.24390243902439024, | |
| "grad_norm": 17.375, | |
| "learning_rate": 4.878048780487805e-05, | |
| "loss": 17.5136, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3048780487804878, | |
| "grad_norm": 7.59375, | |
| "learning_rate": 6.097560975609756e-05, | |
| "loss": 16.4183, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.36585365853658536, | |
| "grad_norm": 4.09375, | |
| "learning_rate": 7.317073170731707e-05, | |
| "loss": 15.0038, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4268292682926829, | |
| "grad_norm": 8.25, | |
| "learning_rate": 8.53658536585366e-05, | |
| "loss": 14.2277, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.4878048780487805, | |
| "grad_norm": 16.5, | |
| "learning_rate": 9.75609756097561e-05, | |
| "loss": 13.0434, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5487804878048781, | |
| "grad_norm": 35.25, | |
| "learning_rate": 0.00010975609756097563, | |
| "loss": 8.9174, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6097560975609756, | |
| "grad_norm": 11.125, | |
| "learning_rate": 0.00012195121951219512, | |
| "loss": 2.4207, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6707317073170732, | |
| "grad_norm": 1.5625, | |
| "learning_rate": 0.00013414634146341464, | |
| "loss": 1.5637, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 1.9921875, | |
| "learning_rate": 0.00014634146341463414, | |
| "loss": 1.3611, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7926829268292683, | |
| "grad_norm": 1.5, | |
| "learning_rate": 0.00015853658536585366, | |
| "loss": 1.1715, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.8536585365853658, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 0.0001707317073170732, | |
| "loss": 1.0679, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9146341463414634, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 0.0001829268292682927, | |
| "loss": 0.9835, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 0.0001951219512195122, | |
| "loss": 0.987, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.2807745933532715, | |
| "eval_runtime": 1.1183, | |
| "eval_samples_per_second": 1.788, | |
| "eval_steps_per_second": 0.894, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.0365853658536586, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 0.00019999184556954776, | |
| "loss": 0.887, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.0975609756097562, | |
| "grad_norm": 1.6328125, | |
| "learning_rate": 0.0001999420177550043, | |
| "loss": 0.849, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1585365853658536, | |
| "grad_norm": 4.1875, | |
| "learning_rate": 0.00019984691491033906, | |
| "loss": 0.8057, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 3.34375, | |
| "learning_rate": 0.00019970658011837404, | |
| "loss": 0.8035, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2804878048780488, | |
| "grad_norm": 1.4375, | |
| "learning_rate": 0.00019952107695258992, | |
| "loss": 0.7717, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.3414634146341464, | |
| "grad_norm": 1.3046875, | |
| "learning_rate": 0.00019929048944832638, | |
| "loss": 0.7583, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.4024390243902438, | |
| "grad_norm": 2.265625, | |
| "learning_rate": 0.00019901492206471325, | |
| "loss": 0.7363, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.4634146341463414, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.00019869449963734893, | |
| "loss": 0.7397, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.524390243902439, | |
| "grad_norm": 1.5859375, | |
| "learning_rate": 0.00019832936732174834, | |
| "loss": 0.7268, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.5853658536585367, | |
| "grad_norm": 2.796875, | |
| "learning_rate": 0.00019791969052758562, | |
| "loss": 0.7103, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.6463414634146343, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 0.00019746565484376132, | |
| "loss": 0.7166, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.7073170731707317, | |
| "grad_norm": 1.4296875, | |
| "learning_rate": 0.00019696746595432828, | |
| "loss": 0.7174, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.7682926829268293, | |
| "grad_norm": 2.296875, | |
| "learning_rate": 0.0001964253495453141, | |
| "loss": 0.6822, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.8292682926829267, | |
| "grad_norm": 1.8515625, | |
| "learning_rate": 0.00019583955120248237, | |
| "loss": 0.679, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.8902439024390243, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 0.00019521033630007928, | |
| "loss": 0.6708, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.951219512195122, | |
| "grad_norm": 0.490234375, | |
| "learning_rate": 0.00019453798988061535, | |
| "loss": 0.6859, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1719168424606323, | |
| "eval_runtime": 1.1192, | |
| "eval_samples_per_second": 1.787, | |
| "eval_steps_per_second": 0.893, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 2.0121951219512195, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 0.00019382281652573785, | |
| "loss": 0.6606, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.073170731707317, | |
| "grad_norm": 3.296875, | |
| "learning_rate": 0.00019306514021825118, | |
| "loss": 0.6528, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.1341463414634148, | |
| "grad_norm": 2.609375, | |
| "learning_rate": 0.00019226530419534833, | |
| "loss": 0.6424, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.1951219512195124, | |
| "grad_norm": 1.5703125, | |
| "learning_rate": 0.00019142367079312021, | |
| "loss": 0.6061, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.2560975609756095, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 0.00019054062128241264, | |
| "loss": 0.6186, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.317073170731707, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 0.00018961655569610557, | |
| "loss": 0.6099, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.3780487804878048, | |
| "grad_norm": 0.89453125, | |
| "learning_rate": 0.0001886518926478932, | |
| "loss": 0.6184, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 0.6796875, | |
| "learning_rate": 0.00018764706914264635, | |
| "loss": 0.5949, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.8046875, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 0.5993, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.5609756097560976, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 0.00018551877954036162, | |
| "loss": 0.6041, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.6219512195121952, | |
| "grad_norm": 0.9921875, | |
| "learning_rate": 0.00018439627758611385, | |
| "loss": 0.6005, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.682926829268293, | |
| "grad_norm": 1.0, | |
| "learning_rate": 0.00018323554302364272, | |
| "loss": 0.5917, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.7439024390243905, | |
| "grad_norm": 1.8515625, | |
| "learning_rate": 0.00018203710168075788, | |
| "loss": 0.5894, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.8048780487804876, | |
| "grad_norm": 1.15625, | |
| "learning_rate": 0.0001808014964669293, | |
| "loss": 0.5884, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.8658536585365852, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 0.00017952928712734268, | |
| "loss": 0.5967, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 2.926829268292683, | |
| "grad_norm": 0.3359375, | |
| "learning_rate": 0.00017822104998932713, | |
| "loss": 0.6064, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.9878048780487805, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 0.00017687737770127185, | |
| "loss": 0.5836, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.1479520797729492, | |
| "eval_runtime": 1.1346, | |
| "eval_samples_per_second": 1.763, | |
| "eval_steps_per_second": 0.881, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 3.048780487804878, | |
| "grad_norm": 1.078125, | |
| "learning_rate": 0.00017549887896414851, | |
| "loss": 0.5409, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.1097560975609757, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 0.0001740861782557618, | |
| "loss": 0.5316, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 3.1707317073170733, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 0.0001726399155478529, | |
| "loss": 0.5168, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.231707317073171, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 0.00017116074601618417, | |
| "loss": 0.5243, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 3.292682926829268, | |
| "grad_norm": 0.7890625, | |
| "learning_rate": 0.0001696493397437357, | |
| "loss": 0.5076, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.3536585365853657, | |
| "grad_norm": 0.96875, | |
| "learning_rate": 0.00016810638141714934, | |
| "loss": 0.5251, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 3.4146341463414633, | |
| "grad_norm": 0.890625, | |
| "learning_rate": 0.00016653257001655652, | |
| "loss": 0.5379, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.475609756097561, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 0.0001649286184989315, | |
| "loss": 0.5421, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 3.5365853658536586, | |
| "grad_norm": 0.59765625, | |
| "learning_rate": 0.0001632952534751122, | |
| "loss": 0.5148, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.597560975609756, | |
| "grad_norm": 0.490234375, | |
| "learning_rate": 0.00016163321488063637, | |
| "loss": 0.5222, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 3.658536585365854, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 0.00015994325564054122, | |
| "loss": 0.5229, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.7195121951219514, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 0.00015822614132827837, | |
| "loss": 0.532, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 3.7804878048780486, | |
| "grad_norm": 0.458984375, | |
| "learning_rate": 0.00015648264981889934, | |
| "loss": 0.5401, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.841463414634146, | |
| "grad_norm": 0.453125, | |
| "learning_rate": 0.00015471357093666804, | |
| "loss": 0.5311, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 3.902439024390244, | |
| "grad_norm": 0.396484375, | |
| "learning_rate": 0.00015291970609726007, | |
| "loss": 0.525, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.9634146341463414, | |
| "grad_norm": 0.359375, | |
| "learning_rate": 0.00015110186794471103, | |
| "loss": 0.5178, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.171655535697937, | |
| "eval_runtime": 1.117, | |
| "eval_samples_per_second": 1.791, | |
| "eval_steps_per_second": 0.895, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 4.024390243902439, | |
| "grad_norm": 0.8125, | |
| "learning_rate": 0.00014926087998327837, | |
| "loss": 0.5073, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.085365853658536, | |
| "grad_norm": 0.3671875, | |
| "learning_rate": 0.00014739757620438307, | |
| "loss": 0.4417, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 4.146341463414634, | |
| "grad_norm": 0.7265625, | |
| "learning_rate": 0.0001455128007088009, | |
| "loss": 0.4321, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.2073170731707314, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 0.00014360740732427367, | |
| "loss": 0.4481, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 4.2682926829268295, | |
| "grad_norm": 0.75390625, | |
| "learning_rate": 0.00014168225921871433, | |
| "loss": 0.4489, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.329268292682927, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 0.00013973822850918055, | |
| "loss": 0.4517, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 4.390243902439025, | |
| "grad_norm": 0.78125, | |
| "learning_rate": 0.0001377761958667946, | |
| "loss": 0.4441, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.451219512195122, | |
| "grad_norm": 0.59375, | |
| "learning_rate": 0.00013579705011778766, | |
| "loss": 0.4539, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 4.512195121951219, | |
| "grad_norm": 0.578125, | |
| "learning_rate": 0.00013380168784085027, | |
| "loss": 0.4489, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.573170731707317, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 0.00013179101296097035, | |
| "loss": 0.4443, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 4.634146341463414, | |
| "grad_norm": 0.6796875, | |
| "learning_rate": 0.00012976593633994346, | |
| "loss": 0.4664, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.695121951219512, | |
| "grad_norm": 0.43359375, | |
| "learning_rate": 0.0001277273753637408, | |
| "loss": 0.4572, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 4.7560975609756095, | |
| "grad_norm": 0.84765625, | |
| "learning_rate": 0.00012567625352692127, | |
| "loss": 0.4675, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.817073170731708, | |
| "grad_norm": 0.486328125, | |
| "learning_rate": 0.0001236135000142765, | |
| "loss": 0.45, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 4.878048780487805, | |
| "grad_norm": 0.4140625, | |
| "learning_rate": 0.00012154004927989815, | |
| "loss": 0.4538, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.939024390243903, | |
| "grad_norm": 0.3984375, | |
| "learning_rate": 0.00011945684062385803, | |
| "loss": 0.4565, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.4140625, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.4668, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.2044252157211304, | |
| "eval_runtime": 1.1171, | |
| "eval_samples_per_second": 1.79, | |
| "eval_steps_per_second": 0.895, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 5.060975609756097, | |
| "grad_norm": 0.63671875, | |
| "learning_rate": 0.00011526492842188745, | |
| "loss": 0.3794, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 5.121951219512195, | |
| "grad_norm": 0.83984375, | |
| "learning_rate": 0.0001131581238665465, | |
| "loss": 0.376, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.182926829268292, | |
| "grad_norm": 0.453125, | |
| "learning_rate": 0.00011104535851045539, | |
| "loss": 0.3721, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 5.2439024390243905, | |
| "grad_norm": 0.470703125, | |
| "learning_rate": 0.00010892758946371944, | |
| "loss": 0.3812, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.304878048780488, | |
| "grad_norm": 0.78125, | |
| "learning_rate": 0.00010680577610318072, | |
| "loss": 0.3748, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 5.365853658536586, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 0.00010468087963780789, | |
| "loss": 0.3698, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.426829268292683, | |
| "grad_norm": 0.48828125, | |
| "learning_rate": 0.00010255386267325602, | |
| "loss": 0.3849, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 5.487804878048781, | |
| "grad_norm": 0.45703125, | |
| "learning_rate": 0.00010042568877579388, | |
| "loss": 0.3831, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.548780487804878, | |
| "grad_norm": 0.470703125, | |
| "learning_rate": 9.829732203579584e-05, | |
| "loss": 0.3765, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 5.609756097560975, | |
| "grad_norm": 0.58984375, | |
| "learning_rate": 9.616972663099647e-05, | |
| "loss": 0.4006, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.670731707317073, | |
| "grad_norm": 0.4609375, | |
| "learning_rate": 9.404386638970542e-05, | |
| "loss": 0.384, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 5.7317073170731705, | |
| "grad_norm": 0.41796875, | |
| "learning_rate": 9.192070435418079e-05, | |
| "loss": 0.3804, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.7926829268292686, | |
| "grad_norm": 0.56640625, | |
| "learning_rate": 8.980120234435849e-05, | |
| "loss": 0.39, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 5.853658536585366, | |
| "grad_norm": 0.7265625, | |
| "learning_rate": 8.768632052213531e-05, | |
| "loss": 0.3881, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.914634146341464, | |
| "grad_norm": 0.44921875, | |
| "learning_rate": 8.557701695640321e-05, | |
| "loss": 0.3908, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 5.975609756097561, | |
| "grad_norm": 0.4140625, | |
| "learning_rate": 8.347424718903151e-05, | |
| "loss": 0.3955, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.325190782546997, | |
| "eval_runtime": 1.1255, | |
| "eval_samples_per_second": 1.777, | |
| "eval_steps_per_second": 0.888, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 6.036585365853658, | |
| "grad_norm": 0.3984375, | |
| "learning_rate": 8.13789638019942e-05, | |
| "loss": 0.3366, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 6.097560975609756, | |
| "grad_norm": 0.408203125, | |
| "learning_rate": 7.929211598583794e-05, | |
| "loss": 0.3141, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.158536585365853, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 7.721464910968627e-05, | |
| "loss": 0.3234, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 6.219512195121951, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 7.514750429297528e-05, | |
| "loss": 0.3232, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 6.280487804878049, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 7.309161797911441e-05, | |
| "loss": 0.3198, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 6.341463414634147, | |
| "grad_norm": 0.43359375, | |
| "learning_rate": 7.104792151126515e-05, | |
| "loss": 0.3236, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 6.402439024390244, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 6.901734071043071e-05, | |
| "loss": 0.3126, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 6.463414634146342, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 6.700079545604708e-05, | |
| "loss": 0.3167, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.524390243902439, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 6.499919926926566e-05, | |
| "loss": 0.3412, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 6.585365853658536, | |
| "grad_norm": 0.5703125, | |
| "learning_rate": 6.301345889911637e-05, | |
| "loss": 0.3269, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.646341463414634, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 6.104447391173858e-05, | |
| "loss": 0.3309, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 6.7073170731707314, | |
| "grad_norm": 0.482421875, | |
| "learning_rate": 5.909313628286601e-05, | |
| "loss": 0.3172, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.7682926829268295, | |
| "grad_norm": 0.53125, | |
| "learning_rate": 5.716032999375006e-05, | |
| "loss": 0.326, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 6.829268292682927, | |
| "grad_norm": 0.439453125, | |
| "learning_rate": 5.524693063070492e-05, | |
| "loss": 0.3207, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.890243902439025, | |
| "grad_norm": 0.55859375, | |
| "learning_rate": 5.335380498845559e-05, | |
| "loss": 0.3295, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 6.951219512195122, | |
| "grad_norm": 0.47265625, | |
| "learning_rate": 5.148181067746862e-05, | |
| "loss": 0.3233, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.4225410223007202, | |
| "eval_runtime": 1.1188, | |
| "eval_samples_per_second": 1.788, | |
| "eval_steps_per_second": 0.894, | |
| "step": 574 | |
| }, | |
| { | |
| "epoch": 7.012195121951219, | |
| "grad_norm": 0.421875, | |
| "learning_rate": 4.963179573544357e-05, | |
| "loss": 0.3129, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 7.073170731707317, | |
| "grad_norm": 0.57421875, | |
| "learning_rate": 4.7804598243140666e-05, | |
| "loss": 0.2764, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 7.134146341463414, | |
| "grad_norm": 0.5, | |
| "learning_rate": 4.60010459447196e-05, | |
| "loss": 0.2792, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 7.195121951219512, | |
| "grad_norm": 0.486328125, | |
| "learning_rate": 4.422195587276058e-05, | |
| "loss": 0.2799, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 7.2560975609756095, | |
| "grad_norm": 0.47265625, | |
| "learning_rate": 4.2468133978137945e-05, | |
| "loss": 0.2759, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 7.317073170731708, | |
| "grad_norm": 0.48046875, | |
| "learning_rate": 4.0740374764914136e-05, | |
| "loss": 0.2697, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.378048780487805, | |
| "grad_norm": 0.439453125, | |
| "learning_rate": 3.903946093041877e-05, | |
| "loss": 0.2917, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 7.439024390243903, | |
| "grad_norm": 0.47265625, | |
| "learning_rate": 3.736616301067694e-05, | |
| "loss": 0.2748, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.50390625, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 0.2771, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 7.560975609756097, | |
| "grad_norm": 0.5, | |
| "learning_rate": 3.410543416432069e-05, | |
| "loss": 0.2788, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.621951219512195, | |
| "grad_norm": 0.44921875, | |
| "learning_rate": 3.2519480390159806e-05, | |
| "loss": 0.2829, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 7.682926829268292, | |
| "grad_norm": 0.494140625, | |
| "learning_rate": 3.096409616649023e-05, | |
| "loss": 0.2708, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.7439024390243905, | |
| "grad_norm": 0.46875, | |
| "learning_rate": 2.9439986102536043e-05, | |
| "loss": 0.2945, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 7.804878048780488, | |
| "grad_norm": 0.482421875, | |
| "learning_rate": 2.794784063992131e-05, | |
| "loss": 0.2833, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.865853658536586, | |
| "grad_norm": 0.4375, | |
| "learning_rate": 2.6488335739891178e-05, | |
| "loss": 0.2767, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 7.926829268292683, | |
| "grad_norm": 0.46875, | |
| "learning_rate": 2.50621325770927e-05, | |
| "loss": 0.2725, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.987804878048781, | |
| "grad_norm": 0.4453125, | |
| "learning_rate": 2.366987724005404e-05, | |
| "loss": 0.2669, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.611946702003479, | |
| "eval_runtime": 1.1206, | |
| "eval_samples_per_second": 1.785, | |
| "eval_steps_per_second": 0.892, | |
| "step": 656 | |
| }, | |
| { | |
| "epoch": 8.048780487804878, | |
| "grad_norm": 0.435546875, | |
| "learning_rate": 2.2312200438498043e-05, | |
| "loss": 0.2564, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 8.109756097560975, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 2.0989717217622652e-05, | |
| "loss": 0.2598, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 8.170731707317072, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 1.9703026679477256e-05, | |
| "loss": 0.2507, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 8.231707317073171, | |
| "grad_norm": 0.3984375, | |
| "learning_rate": 1.8452711711561842e-05, | |
| "loss": 0.2433, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 8.292682926829269, | |
| "grad_norm": 0.408203125, | |
| "learning_rate": 1.7239338722771327e-05, | |
| "loss": 0.251, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 8.353658536585366, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 1.6063457386805004e-05, | |
| "loss": 0.252, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 8.414634146341463, | |
| "grad_norm": 0.404296875, | |
| "learning_rate": 1.4925600393157324e-05, | |
| "loss": 0.2471, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 8.475609756097562, | |
| "grad_norm": 0.494140625, | |
| "learning_rate": 1.3826283205802427e-05, | |
| "loss": 0.2526, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 8.536585365853659, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 1.2766003829682505e-05, | |
| "loss": 0.2517, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.597560975609756, | |
| "grad_norm": 0.4375, | |
| "learning_rate": 1.1745242585104955e-05, | |
| "loss": 0.2479, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 8.658536585365853, | |
| "grad_norm": 0.404296875, | |
| "learning_rate": 1.0764461890151112e-05, | |
| "loss": 0.2485, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 8.71951219512195, | |
| "grad_norm": 0.44140625, | |
| "learning_rate": 9.824106051194859e-06, | |
| "loss": 0.2559, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 8.78048780487805, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 8.924601061626048e-06, | |
| "loss": 0.2588, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.841463414634147, | |
| "grad_norm": 0.40625, | |
| "learning_rate": 8.066354408870048e-06, | |
| "loss": 0.2518, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 8.902439024390244, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 7.249754889790539e-06, | |
| "loss": 0.2542, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 8.963414634146341, | |
| "grad_norm": 0.453125, | |
| "learning_rate": 6.475172434559573e-06, | |
| "loss": 0.2591, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.73529052734375, | |
| "eval_runtime": 1.1218, | |
| "eval_samples_per_second": 1.783, | |
| "eval_steps_per_second": 0.891, | |
| "step": 738 | |
| }, | |
| { | |
| "epoch": 9.024390243902438, | |
| "grad_norm": 0.419921875, | |
| "learning_rate": 5.742957939074412e-06, | |
| "loss": 0.2467, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 9.085365853658537, | |
| "grad_norm": 0.384765625, | |
| "learning_rate": 5.0534431059970685e-06, | |
| "loss": 0.2449, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 9.146341463414634, | |
| "grad_norm": 0.38671875, | |
| "learning_rate": 4.40694029448877e-06, | |
| "loss": 0.244, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 9.207317073170731, | |
| "grad_norm": 0.419921875, | |
| "learning_rate": 3.803742378707198e-06, | |
| "loss": 0.2612, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 9.268292682926829, | |
| "grad_norm": 0.384765625, | |
| "learning_rate": 3.2441226151306404e-06, | |
| "loss": 0.2497, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 9.329268292682928, | |
| "grad_norm": 0.41796875, | |
| "learning_rate": 2.7283345187693264e-06, | |
| "loss": 0.2571, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 9.390243902439025, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 2.256611748319792e-06, | |
| "loss": 0.2458, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 9.451219512195122, | |
| "grad_norm": 0.427734375, | |
| "learning_rate": 1.8291680003145073e-06, | |
| "loss": 0.2585, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 9.512195121951219, | |
| "grad_norm": 0.423828125, | |
| "learning_rate": 1.4461969123145457e-06, | |
| "loss": 0.2516, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 9.573170731707316, | |
| "grad_norm": 0.39453125, | |
| "learning_rate": 1.107871975189234e-06, | |
| "loss": 0.2396, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 9.634146341463415, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 8.143464545226298e-07, | |
| "loss": 0.2441, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 9.695121951219512, | |
| "grad_norm": 0.3984375, | |
| "learning_rate": 5.657533211820942e-07, | |
| "loss": 0.2417, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 9.75609756097561, | |
| "grad_norm": 0.400390625, | |
| "learning_rate": 3.622051910808666e-07, | |
| "loss": 0.2425, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.817073170731707, | |
| "grad_norm": 0.412109375, | |
| "learning_rate": 2.037942741615617e-07, | |
| "loss": 0.2447, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 9.878048780487806, | |
| "grad_norm": 0.39453125, | |
| "learning_rate": 9.059233262386225e-08, | |
| "loss": 0.2512, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 9.939024390243903, | |
| "grad_norm": 0.39453125, | |
| "learning_rate": 2.2650648415334376e-08, | |
| "loss": 0.2365, | |
| "step": 815 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.3828125, | |
| "learning_rate": 0.0, | |
| "loss": 0.2367, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.750213384628296, | |
| "eval_runtime": 1.1197, | |
| "eval_samples_per_second": 1.786, | |
| "eval_steps_per_second": 0.893, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 820, | |
| "total_flos": 1.257706785410646e+18, | |
| "train_loss": 1.5742560074096772, | |
| "train_runtime": 5911.531, | |
| "train_samples_per_second": 2.214, | |
| "train_steps_per_second": 0.139 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 820, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 1.257706785410646e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |