| { |
| "best_global_step": 360, |
| "best_metric": 0.07681839168071747, |
| "best_model_checkpoint": "saves_stability/prompt-tuning/llama-3-8b-instruct/train_copa_1757340277/checkpoint-360", |
| "epoch": 10.0, |
| "eval_steps": 45, |
| "global_step": 900, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.05555555555555555, |
| "grad_norm": 10.1875, |
| "learning_rate": 2.2222222222222225e-06, |
| "loss": 0.2556, |
| "num_input_tokens_seen": 1600, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1111111111111111, |
| "grad_norm": 11.4375, |
| "learning_rate": 5e-06, |
| "loss": 0.388, |
| "num_input_tokens_seen": 3200, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.16666666666666666, |
| "grad_norm": 13.625, |
| "learning_rate": 7.777777777777777e-06, |
| "loss": 0.3683, |
| "num_input_tokens_seen": 4736, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.0555555555555555e-05, |
| "loss": 0.2387, |
| "num_input_tokens_seen": 6368, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.2777777777777778, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 0.3256, |
| "num_input_tokens_seen": 7872, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 2.125, |
| "learning_rate": 1.6111111111111115e-05, |
| "loss": 0.2292, |
| "num_input_tokens_seen": 9440, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.3888888888888889, |
| "grad_norm": 6.71875, |
| "learning_rate": 1.888888888888889e-05, |
| "loss": 0.171, |
| "num_input_tokens_seen": 11008, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 1.5, |
| "learning_rate": 2.1666666666666667e-05, |
| "loss": 0.1258, |
| "num_input_tokens_seen": 12608, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 2.234375, |
| "learning_rate": 2.4444444444444445e-05, |
| "loss": 0.081, |
| "num_input_tokens_seen": 14144, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 0.12154486030340195, |
| "eval_runtime": 0.894, |
| "eval_samples_per_second": 44.742, |
| "eval_steps_per_second": 11.186, |
| "num_input_tokens_seen": 14144, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5555555555555556, |
| "grad_norm": 0.3984375, |
| "learning_rate": 2.7222222222222223e-05, |
| "loss": 0.3126, |
| "num_input_tokens_seen": 15744, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6111111111111112, |
| "grad_norm": 0.341796875, |
| "learning_rate": 3e-05, |
| "loss": 0.2832, |
| "num_input_tokens_seen": 17312, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 0.255859375, |
| "learning_rate": 3.277777777777778e-05, |
| "loss": 0.0083, |
| "num_input_tokens_seen": 18880, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.7222222222222222, |
| "grad_norm": 1.9921875, |
| "learning_rate": 3.555555555555556e-05, |
| "loss": 0.0213, |
| "num_input_tokens_seen": 20416, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7777777777777778, |
| "grad_norm": 0.12890625, |
| "learning_rate": 3.8333333333333334e-05, |
| "loss": 0.1238, |
| "num_input_tokens_seen": 21920, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 1.65625, |
| "learning_rate": 4.111111111111111e-05, |
| "loss": 0.13, |
| "num_input_tokens_seen": 23488, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 1.359375, |
| "learning_rate": 4.388888888888889e-05, |
| "loss": 0.1664, |
| "num_input_tokens_seen": 25056, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.9444444444444444, |
| "grad_norm": 8.5, |
| "learning_rate": 4.666666666666667e-05, |
| "loss": 0.3319, |
| "num_input_tokens_seen": 26592, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 15.0, |
| "learning_rate": 4.9444444444444446e-05, |
| "loss": 0.3339, |
| "num_input_tokens_seen": 28192, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.10366685688495636, |
| "eval_runtime": 0.8933, |
| "eval_samples_per_second": 44.778, |
| "eval_steps_per_second": 11.195, |
| "num_input_tokens_seen": 28192, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0555555555555556, |
| "grad_norm": 3.640625, |
| "learning_rate": 4.9996991493233693e-05, |
| "loss": 0.1849, |
| "num_input_tokens_seen": 29792, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.263671875, |
| "learning_rate": 4.99847706754774e-05, |
| "loss": 0.0734, |
| "num_input_tokens_seen": 31360, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.1666666666666667, |
| "grad_norm": 0.2080078125, |
| "learning_rate": 4.9963154107272295e-05, |
| "loss": 0.0034, |
| "num_input_tokens_seen": 32832, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.2222222222222223, |
| "grad_norm": 0.04150390625, |
| "learning_rate": 4.993214991772563e-05, |
| "loss": 0.0428, |
| "num_input_tokens_seen": 34336, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.2777777777777777, |
| "grad_norm": 13.625, |
| "learning_rate": 4.989176976624511e-05, |
| "loss": 0.2746, |
| "num_input_tokens_seen": 35936, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 14.9375, |
| "learning_rate": 4.9842028838154285e-05, |
| "loss": 0.1004, |
| "num_input_tokens_seen": 37536, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.3888888888888888, |
| "grad_norm": 2.25, |
| "learning_rate": 4.978294583898196e-05, |
| "loss": 0.0144, |
| "num_input_tokens_seen": 39040, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.4444444444444444, |
| "grad_norm": 0.4765625, |
| "learning_rate": 4.971454298742779e-05, |
| "loss": 0.2119, |
| "num_input_tokens_seen": 40576, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 30.75, |
| "learning_rate": 4.963684600700679e-05, |
| "loss": 0.0463, |
| "num_input_tokens_seen": 42208, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5, |
| "eval_loss": 0.09644050151109695, |
| "eval_runtime": 0.8938, |
| "eval_samples_per_second": 44.751, |
| "eval_steps_per_second": 11.188, |
| "num_input_tokens_seen": 42208, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 16.0, |
| "learning_rate": 4.9549884116375714e-05, |
| "loss": 0.1829, |
| "num_input_tokens_seen": 43776, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.6111111111111112, |
| "grad_norm": 0.267578125, |
| "learning_rate": 4.9453690018345144e-05, |
| "loss": 0.0651, |
| "num_input_tokens_seen": 45376, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 13.1875, |
| "learning_rate": 4.934829988758131e-05, |
| "loss": 0.2118, |
| "num_input_tokens_seen": 46944, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.7222222222222223, |
| "grad_norm": 15.5, |
| "learning_rate": 4.923375335700223e-05, |
| "loss": 0.0687, |
| "num_input_tokens_seen": 48512, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 0.42578125, |
| "learning_rate": 4.9110093502873476e-05, |
| "loss": 0.0971, |
| "num_input_tokens_seen": 50112, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.8333333333333335, |
| "grad_norm": 13.125, |
| "learning_rate": 4.897736682860885e-05, |
| "loss": 0.2739, |
| "num_input_tokens_seen": 51680, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.8888888888888888, |
| "grad_norm": 1.8046875, |
| "learning_rate": 4.883562324728241e-05, |
| "loss": 0.094, |
| "num_input_tokens_seen": 53216, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.9444444444444444, |
| "grad_norm": 1.9453125, |
| "learning_rate": 4.868491606285823e-05, |
| "loss": 0.2348, |
| "num_input_tokens_seen": 54720, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 11.9375, |
| "learning_rate": 4.8525301950144894e-05, |
| "loss": 0.082, |
| "num_input_tokens_seen": 56256, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.07768825441598892, |
| "eval_runtime": 0.9012, |
| "eval_samples_per_second": 44.386, |
| "eval_steps_per_second": 11.097, |
| "num_input_tokens_seen": 56256, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0555555555555554, |
| "grad_norm": 44.75, |
| "learning_rate": 4.835684093348244e-05, |
| "loss": 0.0399, |
| "num_input_tokens_seen": 57856, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.111111111111111, |
| "grad_norm": 8.0, |
| "learning_rate": 4.817959636416969e-05, |
| "loss": 0.0204, |
| "num_input_tokens_seen": 59456, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.1666666666666665, |
| "grad_norm": 5.09375, |
| "learning_rate": 4.7993634896640394e-05, |
| "loss": 0.1529, |
| "num_input_tokens_seen": 61024, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 0.25, |
| "learning_rate": 4.779902646339722e-05, |
| "loss": 0.0281, |
| "num_input_tokens_seen": 62592, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.2777777777777777, |
| "grad_norm": 24.0, |
| "learning_rate": 4.759584424871302e-05, |
| "loss": 0.1464, |
| "num_input_tokens_seen": 64128, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 0.384765625, |
| "learning_rate": 4.7384164661109176e-05, |
| "loss": 0.2188, |
| "num_input_tokens_seen": 65696, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.388888888888889, |
| "grad_norm": 15.875, |
| "learning_rate": 4.7164067304621536e-05, |
| "loss": 0.2274, |
| "num_input_tokens_seen": 67200, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 17.625, |
| "learning_rate": 4.693563494886455e-05, |
| "loss": 0.159, |
| "num_input_tokens_seen": 68800, |
| "step": 220 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 0.359375, |
| "learning_rate": 4.669895349790502e-05, |
| "loss": 0.2678, |
| "num_input_tokens_seen": 70368, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.08221504837274551, |
| "eval_runtime": 0.8966, |
| "eval_samples_per_second": 44.615, |
| "eval_steps_per_second": 11.154, |
| "num_input_tokens_seen": 70368, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5555555555555554, |
| "grad_norm": 23.625, |
| "learning_rate": 4.645411195795709e-05, |
| "loss": 0.0935, |
| "num_input_tokens_seen": 71872, |
| "step": 230 |
| }, |
| { |
| "epoch": 2.611111111111111, |
| "grad_norm": 24.5, |
| "learning_rate": 4.620120240391065e-05, |
| "loss": 0.2316, |
| "num_input_tokens_seen": 73472, |
| "step": 235 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 36.75, |
| "learning_rate": 4.5940319944705736e-05, |
| "loss": 0.0206, |
| "num_input_tokens_seen": 75040, |
| "step": 240 |
| }, |
| { |
| "epoch": 2.7222222222222223, |
| "grad_norm": 2.75, |
| "learning_rate": 4.567156268756594e-05, |
| "loss": 0.0223, |
| "num_input_tokens_seen": 76544, |
| "step": 245 |
| }, |
| { |
| "epoch": 2.7777777777777777, |
| "grad_norm": 5.625, |
| "learning_rate": 4.539503170110431e-05, |
| "loss": 0.0813, |
| "num_input_tokens_seen": 78048, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.8333333333333335, |
| "grad_norm": 38.25, |
| "learning_rate": 4.5110830977315556e-05, |
| "loss": 0.0314, |
| "num_input_tokens_seen": 79680, |
| "step": 255 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 0.045654296875, |
| "learning_rate": 4.4819067392468944e-05, |
| "loss": 0.0165, |
| "num_input_tokens_seen": 81184, |
| "step": 260 |
| }, |
| { |
| "epoch": 2.9444444444444446, |
| "grad_norm": 9.0, |
| "learning_rate": 4.4519850666916484e-05, |
| "loss": 0.0959, |
| "num_input_tokens_seen": 82784, |
| "step": 265 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.93359375, |
| "learning_rate": 4.4213293323831585e-05, |
| "loss": 0.0878, |
| "num_input_tokens_seen": 84320, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.0919964462518692, |
| "eval_runtime": 0.9076, |
| "eval_samples_per_second": 44.073, |
| "eval_steps_per_second": 11.018, |
| "num_input_tokens_seen": 84320, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0555555555555554, |
| "grad_norm": 4.625, |
| "learning_rate": 4.38995106468937e-05, |
| "loss": 0.1456, |
| "num_input_tokens_seen": 85920, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 0.5078125, |
| "learning_rate": 4.357862063693486e-05, |
| "loss": 0.0032, |
| "num_input_tokens_seen": 87456, |
| "step": 280 |
| }, |
| { |
| "epoch": 3.1666666666666665, |
| "grad_norm": 0.020751953125, |
| "learning_rate": 4.325074396756437e-05, |
| "loss": 0.0481, |
| "num_input_tokens_seen": 89056, |
| "step": 285 |
| }, |
| { |
| "epoch": 3.2222222222222223, |
| "grad_norm": 0.40234375, |
| "learning_rate": 4.2916003939788403e-05, |
| "loss": 0.0183, |
| "num_input_tokens_seen": 90560, |
| "step": 290 |
| }, |
| { |
| "epoch": 3.2777777777777777, |
| "grad_norm": 55.25, |
| "learning_rate": 4.257452643564155e-05, |
| "loss": 0.1153, |
| "num_input_tokens_seen": 92128, |
| "step": 295 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.28515625, |
| "learning_rate": 4.22264398708477e-05, |
| "loss": 0.0686, |
| "num_input_tokens_seen": 93728, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.388888888888889, |
| "grad_norm": 0.18359375, |
| "learning_rate": 4.1871875146528195e-05, |
| "loss": 0.0247, |
| "num_input_tokens_seen": 95264, |
| "step": 305 |
| }, |
| { |
| "epoch": 3.4444444444444446, |
| "grad_norm": 0.28515625, |
| "learning_rate": 4.1510965599975196e-05, |
| "loss": 0.1083, |
| "num_input_tokens_seen": 96768, |
| "step": 310 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 39.0, |
| "learning_rate": 4.114384695450906e-05, |
| "loss": 0.1495, |
| "num_input_tokens_seen": 98400, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5, |
| "eval_loss": 0.15322943031787872, |
| "eval_runtime": 0.9022, |
| "eval_samples_per_second": 44.337, |
| "eval_steps_per_second": 11.084, |
| "num_input_tokens_seen": 98400, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 0.49609375, |
| "learning_rate": 4.077065726843828e-05, |
| "loss": 0.0145, |
| "num_input_tokens_seen": 99968, |
| "step": 320 |
| }, |
| { |
| "epoch": 3.611111111111111, |
| "grad_norm": 43.0, |
| "learning_rate": 4.039153688314145e-05, |
| "loss": 0.0936, |
| "num_input_tokens_seen": 101504, |
| "step": 325 |
| }, |
| { |
| "epoch": 3.6666666666666665, |
| "grad_norm": 75.0, |
| "learning_rate": 4.000662837029062e-05, |
| "loss": 0.2716, |
| "num_input_tokens_seen": 103008, |
| "step": 330 |
| }, |
| { |
| "epoch": 3.7222222222222223, |
| "grad_norm": 0.1279296875, |
| "learning_rate": 3.961607647823583e-05, |
| "loss": 0.072, |
| "num_input_tokens_seen": 104608, |
| "step": 335 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 14.1875, |
| "learning_rate": 3.9220028077571295e-05, |
| "loss": 0.0267, |
| "num_input_tokens_seen": 106144, |
| "step": 340 |
| }, |
| { |
| "epoch": 3.8333333333333335, |
| "grad_norm": 0.126953125, |
| "learning_rate": 3.881863210590332e-05, |
| "loss": 0.0145, |
| "num_input_tokens_seen": 107744, |
| "step": 345 |
| }, |
| { |
| "epoch": 3.888888888888889, |
| "grad_norm": 24.5, |
| "learning_rate": 3.841203951184095e-05, |
| "loss": 0.2283, |
| "num_input_tokens_seen": 109280, |
| "step": 350 |
| }, |
| { |
| "epoch": 3.9444444444444446, |
| "grad_norm": 37.5, |
| "learning_rate": 3.8000403198230387e-05, |
| "loss": 0.1573, |
| "num_input_tokens_seen": 110848, |
| "step": 355 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.765625, |
| "learning_rate": 3.75838779646545e-05, |
| "loss": 0.0075, |
| "num_input_tokens_seen": 112416, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.07681839168071747, |
| "eval_runtime": 0.8997, |
| "eval_samples_per_second": 44.46, |
| "eval_steps_per_second": 11.115, |
| "num_input_tokens_seen": 112416, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.055555555555555, |
| "grad_norm": 0.68359375, |
| "learning_rate": 3.7162620449219e-05, |
| "loss": 0.0784, |
| "num_input_tokens_seen": 113984, |
| "step": 365 |
| }, |
| { |
| "epoch": 4.111111111111111, |
| "grad_norm": 0.10546875, |
| "learning_rate": 3.673678906964727e-05, |
| "loss": 0.0976, |
| "num_input_tokens_seen": 115552, |
| "step": 370 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 3.75, |
| "learning_rate": 3.630654396370594e-05, |
| "loss": 0.0455, |
| "num_input_tokens_seen": 117152, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 36.25, |
| "learning_rate": 3.5872046928983626e-05, |
| "loss": 0.0911, |
| "num_input_tokens_seen": 118784, |
| "step": 380 |
| }, |
| { |
| "epoch": 4.277777777777778, |
| "grad_norm": 42.5, |
| "learning_rate": 3.543346136204545e-05, |
| "loss": 0.1415, |
| "num_input_tokens_seen": 120256, |
| "step": 385 |
| }, |
| { |
| "epoch": 4.333333333333333, |
| "grad_norm": 19.75, |
| "learning_rate": 3.499095219698631e-05, |
| "loss": 0.0421, |
| "num_input_tokens_seen": 121824, |
| "step": 390 |
| }, |
| { |
| "epoch": 4.388888888888889, |
| "grad_norm": 0.353515625, |
| "learning_rate": 3.454468584340588e-05, |
| "loss": 0.0395, |
| "num_input_tokens_seen": 123392, |
| "step": 395 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 28.125, |
| "learning_rate": 3.409483012382879e-05, |
| "loss": 0.2481, |
| "num_input_tokens_seen": 124928, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 0.189453125, |
| "learning_rate": 3.364155421059342e-05, |
| "loss": 0.429, |
| "num_input_tokens_seen": 126496, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.5, |
| "eval_loss": 0.15622766315937042, |
| "eval_runtime": 0.8897, |
| "eval_samples_per_second": 44.961, |
| "eval_steps_per_second": 11.24, |
| "num_input_tokens_seen": 126496, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.555555555555555, |
| "grad_norm": 5.625, |
| "learning_rate": 3.318502856223311e-05, |
| "loss": 0.086, |
| "num_input_tokens_seen": 128064, |
| "step": 410 |
| }, |
| { |
| "epoch": 4.611111111111111, |
| "grad_norm": 0.3203125, |
| "learning_rate": 3.272542485937369e-05, |
| "loss": 0.0181, |
| "num_input_tokens_seen": 129632, |
| "step": 415 |
| }, |
| { |
| "epoch": 4.666666666666667, |
| "grad_norm": 1.0234375, |
| "learning_rate": 3.2262915940171376e-05, |
| "loss": 0.0705, |
| "num_input_tokens_seen": 131200, |
| "step": 420 |
| }, |
| { |
| "epoch": 4.722222222222222, |
| "grad_norm": 0.0174560546875, |
| "learning_rate": 3.1797675735315455e-05, |
| "loss": 0.0622, |
| "num_input_tokens_seen": 132800, |
| "step": 425 |
| }, |
| { |
| "epoch": 4.777777777777778, |
| "grad_norm": 0.125, |
| "learning_rate": 3.132987920262005e-05, |
| "loss": 0.0682, |
| "num_input_tokens_seen": 134336, |
| "step": 430 |
| }, |
| { |
| "epoch": 4.833333333333333, |
| "grad_norm": 2.28125, |
| "learning_rate": 3.085970226122962e-05, |
| "loss": 0.0801, |
| "num_input_tokens_seen": 135904, |
| "step": 435 |
| }, |
| { |
| "epoch": 4.888888888888889, |
| "grad_norm": 5.125, |
| "learning_rate": 3.0387321725463e-05, |
| "loss": 0.011, |
| "num_input_tokens_seen": 137504, |
| "step": 440 |
| }, |
| { |
| "epoch": 4.944444444444445, |
| "grad_norm": 0.05224609375, |
| "learning_rate": 2.9912915238320754e-05, |
| "loss": 0.0134, |
| "num_input_tokens_seen": 139040, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.05322265625, |
| "learning_rate": 2.9436661204680882e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 140544, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.1206727996468544, |
| "eval_runtime": 0.8892, |
| "eval_samples_per_second": 44.984, |
| "eval_steps_per_second": 11.246, |
| "num_input_tokens_seen": 140544, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.055555555555555, |
| "grad_norm": 0.193359375, |
| "learning_rate": 2.8958738724208072e-05, |
| "loss": 0.2756, |
| "num_input_tokens_seen": 142176, |
| "step": 455 |
| }, |
| { |
| "epoch": 5.111111111111111, |
| "grad_norm": 0.0244140625, |
| "learning_rate": 2.8479327524001636e-05, |
| "loss": 0.0008, |
| "num_input_tokens_seen": 143776, |
| "step": 460 |
| }, |
| { |
| "epoch": 5.166666666666667, |
| "grad_norm": 45.25, |
| "learning_rate": 2.7998607891007495e-05, |
| "loss": 0.2259, |
| "num_input_tokens_seen": 145312, |
| "step": 465 |
| }, |
| { |
| "epoch": 5.222222222222222, |
| "grad_norm": 34.75, |
| "learning_rate": 2.7516760604219617e-05, |
| "loss": 0.1091, |
| "num_input_tokens_seen": 146880, |
| "step": 470 |
| }, |
| { |
| "epoch": 5.277777777777778, |
| "grad_norm": 70.5, |
| "learning_rate": 2.7033966866696457e-05, |
| "loss": 0.1289, |
| "num_input_tokens_seen": 148416, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "grad_norm": 0.09423828125, |
| "learning_rate": 2.6550408237417885e-05, |
| "loss": 0.0026, |
| "num_input_tokens_seen": 149952, |
| "step": 480 |
| }, |
| { |
| "epoch": 5.388888888888889, |
| "grad_norm": 0.2392578125, |
| "learning_rate": 2.6066266563008267e-05, |
| "loss": 0.0016, |
| "num_input_tokens_seen": 151488, |
| "step": 485 |
| }, |
| { |
| "epoch": 5.444444444444445, |
| "grad_norm": 23.75, |
| "learning_rate": 2.5581723909351406e-05, |
| "loss": 0.0669, |
| "num_input_tokens_seen": 153024, |
| "step": 490 |
| }, |
| { |
| "epoch": 5.5, |
| "grad_norm": 0.263671875, |
| "learning_rate": 2.5096962493123012e-05, |
| "loss": 0.0092, |
| "num_input_tokens_seen": 154592, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.5, |
| "eval_loss": 0.13451813161373138, |
| "eval_runtime": 0.8947, |
| "eval_samples_per_second": 44.71, |
| "eval_steps_per_second": 11.177, |
| "num_input_tokens_seen": 154592, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.555555555555555, |
| "grad_norm": 1.390625, |
| "learning_rate": 2.461216461326642e-05, |
| "loss": 0.158, |
| "num_input_tokens_seen": 156128, |
| "step": 500 |
| }, |
| { |
| "epoch": 5.611111111111111, |
| "grad_norm": 8.6875, |
| "learning_rate": 2.4127512582437485e-05, |
| "loss": 0.0211, |
| "num_input_tokens_seen": 157728, |
| "step": 505 |
| }, |
| { |
| "epoch": 5.666666666666667, |
| "grad_norm": 65.0, |
| "learning_rate": 2.364318865844416e-05, |
| "loss": 0.0811, |
| "num_input_tokens_seen": 159296, |
| "step": 510 |
| }, |
| { |
| "epoch": 5.722222222222222, |
| "grad_norm": 0.0146484375, |
| "learning_rate": 2.3159374975706884e-05, |
| "loss": 0.0355, |
| "num_input_tokens_seen": 160896, |
| "step": 515 |
| }, |
| { |
| "epoch": 5.777777777777778, |
| "grad_norm": 0.205078125, |
| "learning_rate": 2.2676253476765196e-05, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 162400, |
| "step": 520 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 0.08544921875, |
| "learning_rate": 2.2194005843856636e-05, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 163968, |
| "step": 525 |
| }, |
| { |
| "epoch": 5.888888888888889, |
| "grad_norm": 34.75, |
| "learning_rate": 2.1712813430593436e-05, |
| "loss": 0.0939, |
| "num_input_tokens_seen": 165536, |
| "step": 530 |
| }, |
| { |
| "epoch": 5.944444444444445, |
| "grad_norm": 0.01275634765625, |
| "learning_rate": 2.1232857193762924e-05, |
| "loss": 0.2795, |
| "num_input_tokens_seen": 167168, |
| "step": 535 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.0179443359375, |
| "learning_rate": 2.0754317625276983e-05, |
| "loss": 0.002, |
| "num_input_tokens_seen": 168768, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.15241709351539612, |
| "eval_runtime": 0.8946, |
| "eval_samples_per_second": 44.713, |
| "eval_steps_per_second": 11.178, |
| "num_input_tokens_seen": 168768, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.055555555555555, |
| "grad_norm": 0.0145263671875, |
| "learning_rate": 2.02773746842965e-05, |
| "loss": 0.0236, |
| "num_input_tokens_seen": 170336, |
| "step": 545 |
| }, |
| { |
| "epoch": 6.111111111111111, |
| "grad_norm": 0.0252685546875, |
| "learning_rate": 1.980220772955602e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 171904, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.166666666666667, |
| "grad_norm": 0.046630859375, |
| "learning_rate": 1.932899545191433e-05, |
| "loss": 0.0042, |
| "num_input_tokens_seen": 173440, |
| "step": 555 |
| }, |
| { |
| "epoch": 6.222222222222222, |
| "grad_norm": 0.0986328125, |
| "learning_rate": 1.8857915807156092e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 174976, |
| "step": 560 |
| }, |
| { |
| "epoch": 6.277777777777778, |
| "grad_norm": 53.75, |
| "learning_rate": 1.838914594906995e-05, |
| "loss": 0.4482, |
| "num_input_tokens_seen": 176608, |
| "step": 565 |
| }, |
| { |
| "epoch": 6.333333333333333, |
| "grad_norm": 0.197265625, |
| "learning_rate": 1.792286216282824e-05, |
| "loss": 0.0272, |
| "num_input_tokens_seen": 178144, |
| "step": 570 |
| }, |
| { |
| "epoch": 6.388888888888889, |
| "grad_norm": 0.033447265625, |
| "learning_rate": 1.7459239798693364e-05, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 179680, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.444444444444445, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.699845320607571e-05, |
| "loss": 0.0457, |
| "num_input_tokens_seen": 181248, |
| "step": 580 |
| }, |
| { |
| "epoch": 6.5, |
| "grad_norm": 0.0123291015625, |
| "learning_rate": 1.6540675667967974e-05, |
| "loss": 0.0064, |
| "num_input_tokens_seen": 182848, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.5, |
| "eval_loss": 0.16781441867351532, |
| "eval_runtime": 0.8963, |
| "eval_samples_per_second": 44.628, |
| "eval_steps_per_second": 11.157, |
| "num_input_tokens_seen": 182848, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.555555555555555, |
| "grad_norm": 2.515625, |
| "learning_rate": 1.60860793357805e-05, |
| "loss": 0.0186, |
| "num_input_tokens_seen": 184352, |
| "step": 590 |
| }, |
| { |
| "epoch": 6.611111111111111, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.56348351646022e-05, |
| "loss": 0.0017, |
| "num_input_tokens_seen": 185952, |
| "step": 595 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.765625, |
| "learning_rate": 1.5187112848911323e-05, |
| "loss": 0.2829, |
| "num_input_tokens_seen": 187488, |
| "step": 600 |
| }, |
| { |
| "epoch": 6.722222222222222, |
| "grad_norm": 34.5, |
| "learning_rate": 1.47430807587603e-05, |
| "loss": 0.0253, |
| "num_input_tokens_seen": 189056, |
| "step": 605 |
| }, |
| { |
| "epoch": 6.777777777777778, |
| "grad_norm": 60.0, |
| "learning_rate": 1.430290587645865e-05, |
| "loss": 0.1066, |
| "num_input_tokens_seen": 190592, |
| "step": 610 |
| }, |
| { |
| "epoch": 6.833333333333333, |
| "grad_norm": 68.0, |
| "learning_rate": 1.3866753733777765e-05, |
| "loss": 0.1102, |
| "num_input_tokens_seen": 192192, |
| "step": 615 |
| }, |
| { |
| "epoch": 6.888888888888889, |
| "grad_norm": 0.07177734375, |
| "learning_rate": 1.343478834970121e-05, |
| "loss": 0.1103, |
| "num_input_tokens_seen": 193824, |
| "step": 620 |
| }, |
| { |
| "epoch": 6.944444444444445, |
| "grad_norm": 0.2890625, |
| "learning_rate": 1.3007172168743854e-05, |
| "loss": 0.0178, |
| "num_input_tokens_seen": 195392, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 23.25, |
| "learning_rate": 1.2584065999863102e-05, |
| "loss": 0.0449, |
| "num_input_tokens_seen": 196896, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.14474520087242126, |
| "eval_runtime": 0.8974, |
| "eval_samples_per_second": 44.573, |
| "eval_steps_per_second": 11.143, |
| "num_input_tokens_seen": 196896, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.055555555555555, |
| "grad_norm": 51.75, |
| "learning_rate": 1.2165628955985314e-05, |
| "loss": 0.1441, |
| "num_input_tokens_seen": 198496, |
| "step": 635 |
| }, |
| { |
| "epoch": 7.111111111111111, |
| "grad_norm": 0.0172119140625, |
| "learning_rate": 1.175201839416988e-05, |
| "loss": 0.0082, |
| "num_input_tokens_seen": 200064, |
| "step": 640 |
| }, |
| { |
| "epoch": 7.166666666666667, |
| "grad_norm": 41.75, |
| "learning_rate": 1.1343389856433658e-05, |
| "loss": 0.1989, |
| "num_input_tokens_seen": 201600, |
| "step": 645 |
| }, |
| { |
| "epoch": 7.222222222222222, |
| "grad_norm": 0.015625, |
| "learning_rate": 1.0939897011258001e-05, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 203168, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.277777777777778, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.0541691595800337e-05, |
| "loss": 0.0138, |
| "num_input_tokens_seen": 204704, |
| "step": 655 |
| }, |
| { |
| "epoch": 7.333333333333333, |
| "grad_norm": 9.0, |
| "learning_rate": 1.0148923358832022e-05, |
| "loss": 0.0034, |
| "num_input_tokens_seen": 206272, |
| "step": 660 |
| }, |
| { |
| "epoch": 7.388888888888889, |
| "grad_norm": 0.035400390625, |
| "learning_rate": 9.761740004423927e-06, |
| "loss": 0.0717, |
| "num_input_tokens_seen": 207872, |
| "step": 665 |
| }, |
| { |
| "epoch": 7.444444444444445, |
| "grad_norm": 39.0, |
| "learning_rate": 9.380287136401e-06, |
| "loss": 0.0478, |
| "num_input_tokens_seen": 209408, |
| "step": 670 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 40.75, |
| "learning_rate": 9.00470820358663e-06, |
| "loss": 0.1323, |
| "num_input_tokens_seen": 210912, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.5, |
| "eval_loss": 0.16354098916053772, |
| "eval_runtime": 0.8986, |
| "eval_samples_per_second": 44.514, |
| "eval_steps_per_second": 11.129, |
| "num_input_tokens_seen": 210912, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.555555555555555, |
| "grad_norm": 1.15625, |
| "learning_rate": 8.635144445857406e-06, |
| "loss": 0.1422, |
| "num_input_tokens_seen": 212480, |
| "step": 680 |
| }, |
| { |
| "epoch": 7.611111111111111, |
| "grad_norm": 0.01409912109375, |
| "learning_rate": 8.271734841028553e-06, |
| "loss": 0.0008, |
| "num_input_tokens_seen": 214016, |
| "step": 685 |
| }, |
| { |
| "epoch": 7.666666666666667, |
| "grad_norm": 0.376953125, |
| "learning_rate": 7.914616052590071e-06, |
| "loss": 0.0909, |
| "num_input_tokens_seen": 215584, |
| "step": 690 |
| }, |
| { |
| "epoch": 7.722222222222222, |
| "grad_norm": 0.1572265625, |
| "learning_rate": 7.563922378313218e-06, |
| "loss": 0.0192, |
| "num_input_tokens_seen": 217216, |
| "step": 695 |
| }, |
| { |
| "epoch": 7.777777777777778, |
| "grad_norm": 82.0, |
| "learning_rate": 7.219785699746573e-06, |
| "loss": 0.1705, |
| "num_input_tokens_seen": 218752, |
| "step": 700 |
| }, |
| { |
| "epoch": 7.833333333333333, |
| "grad_norm": 1.71875, |
| "learning_rate": 6.882335432620779e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 220256, |
| "step": 705 |
| }, |
| { |
| "epoch": 7.888888888888889, |
| "grad_norm": 52.75, |
| "learning_rate": 6.55169847818059e-06, |
| "loss": 0.102, |
| "num_input_tokens_seen": 221792, |
| "step": 710 |
| }, |
| { |
| "epoch": 7.944444444444445, |
| "grad_norm": 0.54296875, |
| "learning_rate": 6.22799917546252e-06, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 223424, |
| "step": 715 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 0.072265625, |
| "learning_rate": 5.9113592545359945e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 225024, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.22366900742053986, |
| "eval_runtime": 0.895, |
| "eval_samples_per_second": 44.692, |
| "eval_steps_per_second": 11.173, |
| "num_input_tokens_seen": 225024, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.055555555555555, |
| "grad_norm": 0.045166015625, |
| "learning_rate": 5.601897790725643e-06, |
| "loss": 0.0008, |
| "num_input_tokens_seen": 226560, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.11111111111111, |
| "grad_norm": 2.21875, |
| "learning_rate": 5.299731159831953e-06, |
| "loss": 0.017, |
| "num_input_tokens_seen": 228128, |
| "step": 730 |
| }, |
| { |
| "epoch": 8.166666666666666, |
| "grad_norm": 0.046875, |
| "learning_rate": 5.004972994367102e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 229728, |
| "step": 735 |
| }, |
| { |
| "epoch": 8.222222222222221, |
| "grad_norm": 0.40625, |
| "learning_rate": 4.7177341408224e-06, |
| "loss": 0.1064, |
| "num_input_tokens_seen": 231328, |
| "step": 740 |
| }, |
| { |
| "epoch": 8.277777777777779, |
| "grad_norm": 154.0, |
| "learning_rate": 4.438122617983443e-06, |
| "loss": 0.1377, |
| "num_input_tokens_seen": 232896, |
| "step": 745 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.05029296875, |
| "learning_rate": 4.166243576308712e-06, |
| "loss": 0.2193, |
| "num_input_tokens_seen": 234496, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.38888888888889, |
| "grad_norm": 0.0206298828125, |
| "learning_rate": 3.9021992583867325e-06, |
| "loss": 0.2001, |
| "num_input_tokens_seen": 236064, |
| "step": 755 |
| }, |
| { |
| "epoch": 8.444444444444445, |
| "grad_norm": 0.76953125, |
| "learning_rate": 3.6460889604868626e-06, |
| "loss": 0.0519, |
| "num_input_tokens_seen": 237632, |
| "step": 760 |
| }, |
| { |
| "epoch": 8.5, |
| "grad_norm": 15.25, |
| "learning_rate": 3.398008995217988e-06, |
| "loss": 0.0211, |
| "num_input_tokens_seen": 239200, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.5, |
| "eval_loss": 0.20884668827056885, |
| "eval_runtime": 0.894, |
| "eval_samples_per_second": 44.744, |
| "eval_steps_per_second": 11.186, |
| "num_input_tokens_seen": 239200, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.555555555555555, |
| "grad_norm": 19.375, |
| "learning_rate": 3.158052655309332e-06, |
| "loss": 0.0072, |
| "num_input_tokens_seen": 240832, |
| "step": 770 |
| }, |
| { |
| "epoch": 8.61111111111111, |
| "grad_norm": 48.75, |
| "learning_rate": 2.9263101785268254e-06, |
| "loss": 0.2022, |
| "num_input_tokens_seen": 242304, |
| "step": 775 |
| }, |
| { |
| "epoch": 8.666666666666666, |
| "grad_norm": 0.0233154296875, |
| "learning_rate": 2.7028687137384267e-06, |
| "loss": 0.0023, |
| "num_input_tokens_seen": 243840, |
| "step": 780 |
| }, |
| { |
| "epoch": 8.722222222222221, |
| "grad_norm": 0.75390625, |
| "learning_rate": 2.487812288140945e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 245376, |
| "step": 785 |
| }, |
| { |
| "epoch": 8.777777777777779, |
| "grad_norm": 0.08203125, |
| "learning_rate": 2.281221775660894e-06, |
| "loss": 0.0034, |
| "num_input_tokens_seen": 246880, |
| "step": 790 |
| }, |
| { |
| "epoch": 8.833333333333334, |
| "grad_norm": 58.75, |
| "learning_rate": 2.0831748665410765e-06, |
| "loss": 0.092, |
| "num_input_tokens_seen": 248448, |
| "step": 795 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "grad_norm": 0.026123046875, |
| "learning_rate": 1.893746038124497e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 250016, |
| "step": 800 |
| }, |
| { |
| "epoch": 8.944444444444445, |
| "grad_norm": 0.0233154296875, |
| "learning_rate": 1.713006526846439e-06, |
| "loss": 0.0659, |
| "num_input_tokens_seen": 251584, |
| "step": 805 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 0.01495361328125, |
| "learning_rate": 1.541024301445404e-06, |
| "loss": 0.0121, |
| "num_input_tokens_seen": 253152, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.20731568336486816, |
| "eval_runtime": 0.89, |
| "eval_samples_per_second": 44.945, |
| "eval_steps_per_second": 11.236, |
| "num_input_tokens_seen": 253152, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.055555555555555, |
| "grad_norm": 0.12060546875, |
| "learning_rate": 1.3778640374027985e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 254720, |
| "step": 815 |
| }, |
| { |
| "epoch": 9.11111111111111, |
| "grad_norm": 41.25, |
| "learning_rate": 1.2235870926211619e-06, |
| "loss": 0.154, |
| "num_input_tokens_seen": 256288, |
| "step": 820 |
| }, |
| { |
| "epoch": 9.166666666666666, |
| "grad_norm": 0.05078125, |
| "learning_rate": 1.0782514843499653e-06, |
| "loss": 0.3336, |
| "num_input_tokens_seen": 257824, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.222222222222221, |
| "grad_norm": 0.01904296875, |
| "learning_rate": 9.419118673676924e-07, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 259392, |
| "step": 830 |
| }, |
| { |
| "epoch": 9.277777777777779, |
| "grad_norm": 86.0, |
| "learning_rate": 8.146195134284052e-07, |
| "loss": 0.0512, |
| "num_input_tokens_seen": 260896, |
| "step": 835 |
| }, |
| { |
| "epoch": 9.333333333333334, |
| "grad_norm": 0.177734375, |
| "learning_rate": 6.964222919805391e-07, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 262464, |
| "step": 840 |
| }, |
| { |
| "epoch": 9.38888888888889, |
| "grad_norm": 0.033203125, |
| "learning_rate": 5.87364652165176e-07, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 264000, |
| "step": 845 |
| }, |
| { |
| "epoch": 9.444444444444445, |
| "grad_norm": 0.92578125, |
| "learning_rate": 4.874876061005173e-07, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 265536, |
| "step": 850 |
| }, |
| { |
| "epoch": 9.5, |
| "grad_norm": 10.0625, |
| "learning_rate": 3.9682871345891883e-07, |
| "loss": 0.0034, |
| "num_input_tokens_seen": 267040, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.5, |
| "eval_loss": 0.20875152945518494, |
| "eval_runtime": 0.8968, |
| "eval_samples_per_second": 44.604, |
| "eval_steps_per_second": 11.151, |
| "num_input_tokens_seen": 267040, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.555555555555555, |
| "grad_norm": 0.04248046875, |
| "learning_rate": 3.1542206734221924e-07, |
| "loss": 0.2541, |
| "num_input_tokens_seen": 268576, |
| "step": 860 |
| }, |
| { |
| "epoch": 9.61111111111111, |
| "grad_norm": 0.271484375, |
| "learning_rate": 2.4329828146074095e-07, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 270144, |
| "step": 865 |
| }, |
| { |
| "epoch": 9.666666666666666, |
| "grad_norm": 0.3046875, |
| "learning_rate": 1.8048447862070718e-07, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 271648, |
| "step": 870 |
| }, |
| { |
| "epoch": 9.722222222222221, |
| "grad_norm": 0.0150146484375, |
| "learning_rate": 1.2700428052447033e-07, |
| "loss": 0.0178, |
| "num_input_tokens_seen": 273280, |
| "step": 875 |
| }, |
| { |
| "epoch": 9.777777777777779, |
| "grad_norm": 20.875, |
| "learning_rate": 8.28777988873486e-08, |
| "loss": 0.151, |
| "num_input_tokens_seen": 274880, |
| "step": 880 |
| }, |
| { |
| "epoch": 9.833333333333334, |
| "grad_norm": 0.275390625, |
| "learning_rate": 4.8121627874450625e-08, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 276480, |
| "step": 885 |
| }, |
| { |
| "epoch": 9.88888888888889, |
| "grad_norm": 5.03125, |
| "learning_rate": 2.2748837860270267e-08, |
| "loss": 0.0413, |
| "num_input_tokens_seen": 278080, |
| "step": 890 |
| }, |
| { |
| "epoch": 9.944444444444445, |
| "grad_norm": 15.25, |
| "learning_rate": 6.768970513457151e-09, |
| "loss": 0.0028, |
| "num_input_tokens_seen": 279680, |
| "step": 895 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 83.5, |
| "learning_rate": 1.8803520859811406e-10, |
| "loss": 0.1445, |
| "num_input_tokens_seen": 281312, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.2091631442308426, |
| "eval_runtime": 0.8928, |
| "eval_samples_per_second": 44.803, |
| "eval_steps_per_second": 11.201, |
| "num_input_tokens_seen": 281312, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 281312, |
| "step": 900, |
| "total_flos": 1.2667352215977984e+16, |
| "train_loss": 0.09724792751878138, |
| "train_runtime": 179.6793, |
| "train_samples_per_second": 20.036, |
| "train_steps_per_second": 5.009 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 900, |
| "num_input_tokens_seen": 281312, |
| "num_train_epochs": 10, |
| "save_steps": 45, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.2667352215977984e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
