| { |
| "best_global_step": 90, |
| "best_metric": 0.11805140972137451, |
| "best_model_checkpoint": "saves_stability/lntuning/llama-3-8b-instruct/train_copa_1757340231/checkpoint-90", |
| "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": 12.028129577636719, |
| "learning_rate": 2.2222222222222225e-06, |
| "loss": 0.6551, |
| "num_input_tokens_seen": 1568, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1111111111111111, |
| "grad_norm": 13.552009582519531, |
| "learning_rate": 5e-06, |
| "loss": 0.6673, |
| "num_input_tokens_seen": 3104, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.16666666666666666, |
| "grad_norm": 11.640186309814453, |
| "learning_rate": 7.777777777777777e-06, |
| "loss": 0.9098, |
| "num_input_tokens_seen": 4704, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 9.485243797302246, |
| "learning_rate": 1.0555555555555555e-05, |
| "loss": 0.5073, |
| "num_input_tokens_seen": 6304, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.2777777777777778, |
| "grad_norm": 8.202380180358887, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 0.4478, |
| "num_input_tokens_seen": 7808, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 8.377937316894531, |
| "learning_rate": 1.6111111111111115e-05, |
| "loss": 0.4037, |
| "num_input_tokens_seen": 9344, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.3888888888888889, |
| "grad_norm": 3.752380847930908, |
| "learning_rate": 1.888888888888889e-05, |
| "loss": 0.2965, |
| "num_input_tokens_seen": 10912, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 2.004181385040283, |
| "learning_rate": 2.1666666666666667e-05, |
| "loss": 0.1495, |
| "num_input_tokens_seen": 12512, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 3.194992780685425, |
| "learning_rate": 2.4444444444444445e-05, |
| "loss": 0.2041, |
| "num_input_tokens_seen": 14016, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 0.1784650981426239, |
| "eval_runtime": 0.5749, |
| "eval_samples_per_second": 69.58, |
| "eval_steps_per_second": 17.395, |
| "num_input_tokens_seen": 14016, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5555555555555556, |
| "grad_norm": 1.5340490341186523, |
| "learning_rate": 2.7222222222222223e-05, |
| "loss": 0.0414, |
| "num_input_tokens_seen": 15552, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6111111111111112, |
| "grad_norm": 1.0850095748901367, |
| "learning_rate": 3e-05, |
| "loss": 0.0265, |
| "num_input_tokens_seen": 17120, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 17.816328048706055, |
| "learning_rate": 3.277777777777778e-05, |
| "loss": 0.1182, |
| "num_input_tokens_seen": 18720, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.7222222222222222, |
| "grad_norm": 0.18008385598659515, |
| "learning_rate": 3.555555555555556e-05, |
| "loss": 0.1036, |
| "num_input_tokens_seen": 20288, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7777777777777778, |
| "grad_norm": 0.1256040334701538, |
| "learning_rate": 3.8333333333333334e-05, |
| "loss": 0.021, |
| "num_input_tokens_seen": 21856, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 6.297001361846924, |
| "learning_rate": 4.111111111111111e-05, |
| "loss": 0.1456, |
| "num_input_tokens_seen": 23456, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 6.056818008422852, |
| "learning_rate": 4.388888888888889e-05, |
| "loss": 0.1581, |
| "num_input_tokens_seen": 24992, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.9444444444444444, |
| "grad_norm": 0.5619527101516724, |
| "learning_rate": 4.666666666666667e-05, |
| "loss": 0.1223, |
| "num_input_tokens_seen": 26528, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.8489146828651428, |
| "learning_rate": 4.9444444444444446e-05, |
| "loss": 0.0848, |
| "num_input_tokens_seen": 28096, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.11805140972137451, |
| "eval_runtime": 0.5744, |
| "eval_samples_per_second": 69.634, |
| "eval_steps_per_second": 17.408, |
| "num_input_tokens_seen": 28096, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0555555555555556, |
| "grad_norm": 0.040736857801675797, |
| "learning_rate": 4.9996991493233693e-05, |
| "loss": 0.017, |
| "num_input_tokens_seen": 29632, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.370405912399292, |
| "learning_rate": 4.99847706754774e-05, |
| "loss": 0.0399, |
| "num_input_tokens_seen": 31200, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.1666666666666667, |
| "grad_norm": 0.2983262538909912, |
| "learning_rate": 4.9963154107272295e-05, |
| "loss": 0.0811, |
| "num_input_tokens_seen": 32768, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.2222222222222223, |
| "grad_norm": 0.04243910312652588, |
| "learning_rate": 4.993214991772563e-05, |
| "loss": 0.2488, |
| "num_input_tokens_seen": 34336, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.2777777777777777, |
| "grad_norm": 1.2641880512237549, |
| "learning_rate": 4.989176976624511e-05, |
| "loss": 0.1245, |
| "num_input_tokens_seen": 35872, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 0.5663049221038818, |
| "learning_rate": 4.9842028838154285e-05, |
| "loss": 0.0252, |
| "num_input_tokens_seen": 37440, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.3888888888888888, |
| "grad_norm": 3.6676812171936035, |
| "learning_rate": 4.978294583898196e-05, |
| "loss": 0.0156, |
| "num_input_tokens_seen": 38976, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.4444444444444444, |
| "grad_norm": 5.408911228179932, |
| "learning_rate": 4.971454298742779e-05, |
| "loss": 0.1855, |
| "num_input_tokens_seen": 40544, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 0.3176509439945221, |
| "learning_rate": 4.963684600700679e-05, |
| "loss": 0.069, |
| "num_input_tokens_seen": 42144, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5, |
| "eval_loss": 0.1266736388206482, |
| "eval_runtime": 0.58, |
| "eval_samples_per_second": 68.963, |
| "eval_steps_per_second": 17.241, |
| "num_input_tokens_seen": 42144, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 5.149271488189697, |
| "learning_rate": 4.9549884116375714e-05, |
| "loss": 0.2346, |
| "num_input_tokens_seen": 43744, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.6111111111111112, |
| "grad_norm": 0.3459404408931732, |
| "learning_rate": 4.9453690018345144e-05, |
| "loss": 0.0473, |
| "num_input_tokens_seen": 45280, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 0.04379900172352791, |
| "learning_rate": 4.934829988758131e-05, |
| "loss": 0.1136, |
| "num_input_tokens_seen": 46880, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.7222222222222223, |
| "grad_norm": 6.0745720863342285, |
| "learning_rate": 4.923375335700223e-05, |
| "loss": 0.2481, |
| "num_input_tokens_seen": 48480, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 0.06044352427124977, |
| "learning_rate": 4.9110093502873476e-05, |
| "loss": 0.017, |
| "num_input_tokens_seen": 50080, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.8333333333333335, |
| "grad_norm": 4.317785263061523, |
| "learning_rate": 4.897736682860885e-05, |
| "loss": 0.0766, |
| "num_input_tokens_seen": 51648, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.8888888888888888, |
| "grad_norm": 2.0243148803710938, |
| "learning_rate": 4.883562324728241e-05, |
| "loss": 0.0453, |
| "num_input_tokens_seen": 53216, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.9444444444444444, |
| "grad_norm": 0.2717069983482361, |
| "learning_rate": 4.868491606285823e-05, |
| "loss": 0.0625, |
| "num_input_tokens_seen": 54752, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.03594539687037468, |
| "learning_rate": 4.8525301950144894e-05, |
| "loss": 0.0283, |
| "num_input_tokens_seen": 56352, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.11958863586187363, |
| "eval_runtime": 0.5774, |
| "eval_samples_per_second": 69.278, |
| "eval_steps_per_second": 17.319, |
| "num_input_tokens_seen": 56352, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0555555555555554, |
| "grad_norm": 0.019619405269622803, |
| "learning_rate": 4.835684093348244e-05, |
| "loss": 0.0367, |
| "num_input_tokens_seen": 57888, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.111111111111111, |
| "grad_norm": 0.5464202761650085, |
| "learning_rate": 4.817959636416969e-05, |
| "loss": 0.0118, |
| "num_input_tokens_seen": 59488, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.1666666666666665, |
| "grad_norm": 0.048002611845731735, |
| "learning_rate": 4.7993634896640394e-05, |
| "loss": 0.091, |
| "num_input_tokens_seen": 61024, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 0.032152995467185974, |
| "learning_rate": 4.779902646339722e-05, |
| "loss": 0.051, |
| "num_input_tokens_seen": 62528, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.2777777777777777, |
| "grad_norm": 6.827496528625488, |
| "learning_rate": 4.759584424871302e-05, |
| "loss": 0.1105, |
| "num_input_tokens_seen": 64064, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 0.04550907015800476, |
| "learning_rate": 4.7384164661109176e-05, |
| "loss": 0.1496, |
| "num_input_tokens_seen": 65696, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.388888888888889, |
| "grad_norm": 2.4212865829467773, |
| "learning_rate": 4.7164067304621536e-05, |
| "loss": 0.0791, |
| "num_input_tokens_seen": 67232, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 2.493133544921875, |
| "learning_rate": 4.693563494886455e-05, |
| "loss": 0.0206, |
| "num_input_tokens_seen": 68832, |
| "step": 220 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 4.306882858276367, |
| "learning_rate": 4.669895349790502e-05, |
| "loss": 0.1055, |
| "num_input_tokens_seen": 70432, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.11920864880084991, |
| "eval_runtime": 0.583, |
| "eval_samples_per_second": 68.608, |
| "eval_steps_per_second": 17.152, |
| "num_input_tokens_seen": 70432, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5555555555555554, |
| "grad_norm": 0.6485729813575745, |
| "learning_rate": 4.645411195795709e-05, |
| "loss": 0.0058, |
| "num_input_tokens_seen": 72032, |
| "step": 230 |
| }, |
| { |
| "epoch": 2.611111111111111, |
| "grad_norm": 0.7758520245552063, |
| "learning_rate": 4.620120240391065e-05, |
| "loss": 0.015, |
| "num_input_tokens_seen": 73632, |
| "step": 235 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 0.016528239473700523, |
| "learning_rate": 4.5940319944705736e-05, |
| "loss": 0.1068, |
| "num_input_tokens_seen": 75168, |
| "step": 240 |
| }, |
| { |
| "epoch": 2.7222222222222223, |
| "grad_norm": 2.0096018314361572, |
| "learning_rate": 4.567156268756594e-05, |
| "loss": 0.0228, |
| "num_input_tokens_seen": 76768, |
| "step": 245 |
| }, |
| { |
| "epoch": 2.7777777777777777, |
| "grad_norm": 0.016257042065262794, |
| "learning_rate": 4.539503170110431e-05, |
| "loss": 0.0078, |
| "num_input_tokens_seen": 78336, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.8333333333333335, |
| "grad_norm": 0.037896547466516495, |
| "learning_rate": 4.5110830977315556e-05, |
| "loss": 0.0148, |
| "num_input_tokens_seen": 79872, |
| "step": 255 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 1.2489393949508667, |
| "learning_rate": 4.4819067392468944e-05, |
| "loss": 0.06, |
| "num_input_tokens_seen": 81440, |
| "step": 260 |
| }, |
| { |
| "epoch": 2.9444444444444446, |
| "grad_norm": 0.04921220242977142, |
| "learning_rate": 4.4519850666916484e-05, |
| "loss": 0.0711, |
| "num_input_tokens_seen": 83008, |
| "step": 265 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.06469002366065979, |
| "learning_rate": 4.4213293323831585e-05, |
| "loss": 0.2017, |
| "num_input_tokens_seen": 84544, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.1241941899061203, |
| "eval_runtime": 0.5764, |
| "eval_samples_per_second": 69.395, |
| "eval_steps_per_second": 17.349, |
| "num_input_tokens_seen": 84544, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0555555555555554, |
| "grad_norm": 1.0416898727416992, |
| "learning_rate": 4.38995106468937e-05, |
| "loss": 0.0423, |
| "num_input_tokens_seen": 86112, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 0.030291926115751266, |
| "learning_rate": 4.357862063693486e-05, |
| "loss": 0.1031, |
| "num_input_tokens_seen": 87680, |
| "step": 280 |
| }, |
| { |
| "epoch": 3.1666666666666665, |
| "grad_norm": 0.34328025579452515, |
| "learning_rate": 4.325074396756437e-05, |
| "loss": 0.0353, |
| "num_input_tokens_seen": 89216, |
| "step": 285 |
| }, |
| { |
| "epoch": 3.2222222222222223, |
| "grad_norm": 0.014772674068808556, |
| "learning_rate": 4.2916003939788403e-05, |
| "loss": 0.0296, |
| "num_input_tokens_seen": 90816, |
| "step": 290 |
| }, |
| { |
| "epoch": 3.2777777777777777, |
| "grad_norm": 1.0737226009368896, |
| "learning_rate": 4.257452643564155e-05, |
| "loss": 0.051, |
| "num_input_tokens_seen": 92448, |
| "step": 295 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.3188193440437317, |
| "learning_rate": 4.22264398708477e-05, |
| "loss": 0.0048, |
| "num_input_tokens_seen": 93984, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.388888888888889, |
| "grad_norm": 0.01848280057311058, |
| "learning_rate": 4.1871875146528195e-05, |
| "loss": 0.0704, |
| "num_input_tokens_seen": 95584, |
| "step": 305 |
| }, |
| { |
| "epoch": 3.4444444444444446, |
| "grad_norm": 0.02418445609509945, |
| "learning_rate": 4.1510965599975196e-05, |
| "loss": 0.0083, |
| "num_input_tokens_seen": 97088, |
| "step": 310 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 0.012784923426806927, |
| "learning_rate": 4.114384695450906e-05, |
| "loss": 0.001, |
| "num_input_tokens_seen": 98688, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5, |
| "eval_loss": 0.12571492791175842, |
| "eval_runtime": 0.5841, |
| "eval_samples_per_second": 68.486, |
| "eval_steps_per_second": 17.122, |
| "num_input_tokens_seen": 98688, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 0.1880708634853363, |
| "learning_rate": 4.077065726843828e-05, |
| "loss": 0.1335, |
| "num_input_tokens_seen": 100288, |
| "step": 320 |
| }, |
| { |
| "epoch": 3.611111111111111, |
| "grad_norm": 0.010126232169568539, |
| "learning_rate": 4.039153688314145e-05, |
| "loss": 0.0038, |
| "num_input_tokens_seen": 101824, |
| "step": 325 |
| }, |
| { |
| "epoch": 3.6666666666666665, |
| "grad_norm": 0.01738058216869831, |
| "learning_rate": 4.000662837029062e-05, |
| "loss": 0.0093, |
| "num_input_tokens_seen": 103392, |
| "step": 330 |
| }, |
| { |
| "epoch": 3.7222222222222223, |
| "grad_norm": 0.39928749203681946, |
| "learning_rate": 3.961607647823583e-05, |
| "loss": 0.1068, |
| "num_input_tokens_seen": 104928, |
| "step": 335 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 3.9759440422058105, |
| "learning_rate": 3.9220028077571295e-05, |
| "loss": 0.1346, |
| "num_input_tokens_seen": 106560, |
| "step": 340 |
| }, |
| { |
| "epoch": 3.8333333333333335, |
| "grad_norm": 0.045272424817085266, |
| "learning_rate": 3.881863210590332e-05, |
| "loss": 0.0017, |
| "num_input_tokens_seen": 108096, |
| "step": 345 |
| }, |
| { |
| "epoch": 3.888888888888889, |
| "grad_norm": 1.4312611818313599, |
| "learning_rate": 3.841203951184095e-05, |
| "loss": 0.0963, |
| "num_input_tokens_seen": 109664, |
| "step": 350 |
| }, |
| { |
| "epoch": 3.9444444444444446, |
| "grad_norm": 0.02004213072359562, |
| "learning_rate": 3.8000403198230387e-05, |
| "loss": 0.0429, |
| "num_input_tokens_seen": 111232, |
| "step": 355 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.017298046499490738, |
| "learning_rate": 3.75838779646545e-05, |
| "loss": 0.0121, |
| "num_input_tokens_seen": 112800, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.1298300325870514, |
| "eval_runtime": 0.5773, |
| "eval_samples_per_second": 69.285, |
| "eval_steps_per_second": 17.321, |
| "num_input_tokens_seen": 112800, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.055555555555555, |
| "grad_norm": 0.04087962582707405, |
| "learning_rate": 3.7162620449219e-05, |
| "loss": 0.1289, |
| "num_input_tokens_seen": 114400, |
| "step": 365 |
| }, |
| { |
| "epoch": 4.111111111111111, |
| "grad_norm": 6.873142719268799, |
| "learning_rate": 3.673678906964727e-05, |
| "loss": 0.1789, |
| "num_input_tokens_seen": 115936, |
| "step": 370 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 0.16114553809165955, |
| "learning_rate": 3.630654396370594e-05, |
| "loss": 0.0099, |
| "num_input_tokens_seen": 117472, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 0.5829356908798218, |
| "learning_rate": 3.5872046928983626e-05, |
| "loss": 0.1038, |
| "num_input_tokens_seen": 119040, |
| "step": 380 |
| }, |
| { |
| "epoch": 4.277777777777778, |
| "grad_norm": 0.7600432634353638, |
| "learning_rate": 3.543346136204545e-05, |
| "loss": 0.0187, |
| "num_input_tokens_seen": 120640, |
| "step": 385 |
| }, |
| { |
| "epoch": 4.333333333333333, |
| "grad_norm": 0.151974618434906, |
| "learning_rate": 3.499095219698631e-05, |
| "loss": 0.0012, |
| "num_input_tokens_seen": 122176, |
| "step": 390 |
| }, |
| { |
| "epoch": 4.388888888888889, |
| "grad_norm": 0.2258666604757309, |
| "learning_rate": 3.454468584340588e-05, |
| "loss": 0.0239, |
| "num_input_tokens_seen": 123744, |
| "step": 395 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 0.018906690180301666, |
| "learning_rate": 3.409483012382879e-05, |
| "loss": 0.0231, |
| "num_input_tokens_seen": 125280, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 0.05237025022506714, |
| "learning_rate": 3.364155421059342e-05, |
| "loss": 0.0068, |
| "num_input_tokens_seen": 126816, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.5, |
| "eval_loss": 0.13041964173316956, |
| "eval_runtime": 0.5753, |
| "eval_samples_per_second": 69.532, |
| "eval_steps_per_second": 17.383, |
| "num_input_tokens_seen": 126816, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.555555555555555, |
| "grad_norm": 0.11989942193031311, |
| "learning_rate": 3.318502856223311e-05, |
| "loss": 0.0047, |
| "num_input_tokens_seen": 128352, |
| "step": 410 |
| }, |
| { |
| "epoch": 4.611111111111111, |
| "grad_norm": 0.015702011063694954, |
| "learning_rate": 3.272542485937369e-05, |
| "loss": 0.0029, |
| "num_input_tokens_seen": 129920, |
| "step": 415 |
| }, |
| { |
| "epoch": 4.666666666666667, |
| "grad_norm": 0.028754986822605133, |
| "learning_rate": 3.2262915940171376e-05, |
| "loss": 0.02, |
| "num_input_tokens_seen": 131424, |
| "step": 420 |
| }, |
| { |
| "epoch": 4.722222222222222, |
| "grad_norm": 0.010241667740046978, |
| "learning_rate": 3.1797675735315455e-05, |
| "loss": 0.004, |
| "num_input_tokens_seen": 132960, |
| "step": 425 |
| }, |
| { |
| "epoch": 4.777777777777778, |
| "grad_norm": 0.011341402307152748, |
| "learning_rate": 3.132987920262005e-05, |
| "loss": 0.0012, |
| "num_input_tokens_seen": 134528, |
| "step": 430 |
| }, |
| { |
| "epoch": 4.833333333333333, |
| "grad_norm": 0.009936963208019733, |
| "learning_rate": 3.085970226122962e-05, |
| "loss": 0.0168, |
| "num_input_tokens_seen": 136096, |
| "step": 435 |
| }, |
| { |
| "epoch": 4.888888888888889, |
| "grad_norm": 0.03836958855390549, |
| "learning_rate": 3.0387321725463e-05, |
| "loss": 0.0643, |
| "num_input_tokens_seen": 137728, |
| "step": 440 |
| }, |
| { |
| "epoch": 4.944444444444445, |
| "grad_norm": 3.9874274730682373, |
| "learning_rate": 2.9912915238320754e-05, |
| "loss": 0.0452, |
| "num_input_tokens_seen": 139264, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.8843461275100708, |
| "learning_rate": 2.9436661204680882e-05, |
| "loss": 0.0379, |
| "num_input_tokens_seen": 140800, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.1350220888853073, |
| "eval_runtime": 0.5763, |
| "eval_samples_per_second": 69.414, |
| "eval_steps_per_second": 17.353, |
| "num_input_tokens_seen": 140800, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.055555555555555, |
| "grad_norm": 0.2792241871356964, |
| "learning_rate": 2.8958738724208072e-05, |
| "loss": 0.0302, |
| "num_input_tokens_seen": 142400, |
| "step": 455 |
| }, |
| { |
| "epoch": 5.111111111111111, |
| "grad_norm": 0.027824169024825096, |
| "learning_rate": 2.8479327524001636e-05, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 143872, |
| "step": 460 |
| }, |
| { |
| "epoch": 5.166666666666667, |
| "grad_norm": 0.005916323512792587, |
| "learning_rate": 2.7998607891007495e-05, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 145440, |
| "step": 465 |
| }, |
| { |
| "epoch": 5.222222222222222, |
| "grad_norm": 0.008561103604733944, |
| "learning_rate": 2.7516760604219617e-05, |
| "loss": 0.0025, |
| "num_input_tokens_seen": 147040, |
| "step": 470 |
| }, |
| { |
| "epoch": 5.277777777777778, |
| "grad_norm": 0.013526272028684616, |
| "learning_rate": 2.7033966866696457e-05, |
| "loss": 0.0771, |
| "num_input_tokens_seen": 148576, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "grad_norm": 0.13795623183250427, |
| "learning_rate": 2.6550408237417885e-05, |
| "loss": 0.0053, |
| "num_input_tokens_seen": 150112, |
| "step": 480 |
| }, |
| { |
| "epoch": 5.388888888888889, |
| "grad_norm": 0.8442857265472412, |
| "learning_rate": 2.6066266563008267e-05, |
| "loss": 0.0898, |
| "num_input_tokens_seen": 151712, |
| "step": 485 |
| }, |
| { |
| "epoch": 5.444444444444445, |
| "grad_norm": 2.5526673793792725, |
| "learning_rate": 2.5581723909351406e-05, |
| "loss": 0.0661, |
| "num_input_tokens_seen": 153280, |
| "step": 490 |
| }, |
| { |
| "epoch": 5.5, |
| "grad_norm": 0.05172133445739746, |
| "learning_rate": 2.5096962493123012e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 154848, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.5, |
| "eval_loss": 0.13754430413246155, |
| "eval_runtime": 0.58, |
| "eval_samples_per_second": 68.968, |
| "eval_steps_per_second": 17.242, |
| "num_input_tokens_seen": 154848, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.555555555555555, |
| "grad_norm": 0.007516528945416212, |
| "learning_rate": 2.461216461326642e-05, |
| "loss": 0.007, |
| "num_input_tokens_seen": 156352, |
| "step": 500 |
| }, |
| { |
| "epoch": 5.611111111111111, |
| "grad_norm": 3.065181016921997, |
| "learning_rate": 2.4127512582437485e-05, |
| "loss": 0.0205, |
| "num_input_tokens_seen": 157920, |
| "step": 505 |
| }, |
| { |
| "epoch": 5.666666666666667, |
| "grad_norm": 6.442762851715088, |
| "learning_rate": 2.364318865844416e-05, |
| "loss": 0.0562, |
| "num_input_tokens_seen": 159488, |
| "step": 510 |
| }, |
| { |
| "epoch": 5.722222222222222, |
| "grad_norm": 0.27856680750846863, |
| "learning_rate": 2.3159374975706884e-05, |
| "loss": 0.0033, |
| "num_input_tokens_seen": 161056, |
| "step": 515 |
| }, |
| { |
| "epoch": 5.777777777777778, |
| "grad_norm": 0.3110099732875824, |
| "learning_rate": 2.2676253476765196e-05, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 162624, |
| "step": 520 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 5.475592136383057, |
| "learning_rate": 2.2194005843856636e-05, |
| "loss": 0.0876, |
| "num_input_tokens_seen": 164128, |
| "step": 525 |
| }, |
| { |
| "epoch": 5.888888888888889, |
| "grad_norm": 7.484983444213867, |
| "learning_rate": 2.1712813430593436e-05, |
| "loss": 0.0961, |
| "num_input_tokens_seen": 165664, |
| "step": 530 |
| }, |
| { |
| "epoch": 5.944444444444445, |
| "grad_norm": 0.10913576930761337, |
| "learning_rate": 2.1232857193762924e-05, |
| "loss": 0.0025, |
| "num_input_tokens_seen": 167200, |
| "step": 535 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.027029909193515778, |
| "learning_rate": 2.0754317625276983e-05, |
| "loss": 0.0027, |
| "num_input_tokens_seen": 168736, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.13720272481441498, |
| "eval_runtime": 0.5771, |
| "eval_samples_per_second": 69.312, |
| "eval_steps_per_second": 17.328, |
| "num_input_tokens_seen": 168736, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.055555555555555, |
| "grad_norm": 0.03148474916815758, |
| "learning_rate": 2.02773746842965e-05, |
| "loss": 0.0017, |
| "num_input_tokens_seen": 170304, |
| "step": 545 |
| }, |
| { |
| "epoch": 6.111111111111111, |
| "grad_norm": 0.08897438645362854, |
| "learning_rate": 1.980220772955602e-05, |
| "loss": 0.0135, |
| "num_input_tokens_seen": 171840, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.166666666666667, |
| "grad_norm": 0.05110946670174599, |
| "learning_rate": 1.932899545191433e-05, |
| "loss": 0.0008, |
| "num_input_tokens_seen": 173408, |
| "step": 555 |
| }, |
| { |
| "epoch": 6.222222222222222, |
| "grad_norm": 0.011350942775607109, |
| "learning_rate": 1.8857915807156092e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 174976, |
| "step": 560 |
| }, |
| { |
| "epoch": 6.277777777777778, |
| "grad_norm": 0.009600317105650902, |
| "learning_rate": 1.838914594906995e-05, |
| "loss": 0.0227, |
| "num_input_tokens_seen": 176544, |
| "step": 565 |
| }, |
| { |
| "epoch": 6.333333333333333, |
| "grad_norm": 0.027924560010433197, |
| "learning_rate": 1.792286216282824e-05, |
| "loss": 0.0964, |
| "num_input_tokens_seen": 178112, |
| "step": 570 |
| }, |
| { |
| "epoch": 6.388888888888889, |
| "grad_norm": 0.3235127925872803, |
| "learning_rate": 1.7459239798693364e-05, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 179584, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.444444444444445, |
| "grad_norm": 3.067894458770752, |
| "learning_rate": 1.699845320607571e-05, |
| "loss": 0.022, |
| "num_input_tokens_seen": 181152, |
| "step": 580 |
| }, |
| { |
| "epoch": 6.5, |
| "grad_norm": 0.014973540790379047, |
| "learning_rate": 1.6540675667967974e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 182688, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.5, |
| "eval_loss": 0.14122465252876282, |
| "eval_runtime": 0.5961, |
| "eval_samples_per_second": 67.101, |
| "eval_steps_per_second": 16.775, |
| "num_input_tokens_seen": 182688, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.555555555555555, |
| "grad_norm": 0.009066694416105747, |
| "learning_rate": 1.60860793357805e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 184224, |
| "step": 590 |
| }, |
| { |
| "epoch": 6.611111111111111, |
| "grad_norm": 0.17362192273139954, |
| "learning_rate": 1.56348351646022e-05, |
| "loss": 0.0395, |
| "num_input_tokens_seen": 185792, |
| "step": 595 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.032302968204021454, |
| "learning_rate": 1.5187112848911323e-05, |
| "loss": 0.0417, |
| "num_input_tokens_seen": 187392, |
| "step": 600 |
| }, |
| { |
| "epoch": 6.722222222222222, |
| "grad_norm": 0.6071529388427734, |
| "learning_rate": 1.47430807587603e-05, |
| "loss": 0.0832, |
| "num_input_tokens_seen": 188992, |
| "step": 605 |
| }, |
| { |
| "epoch": 6.777777777777778, |
| "grad_norm": 0.04848611727356911, |
| "learning_rate": 1.430290587645865e-05, |
| "loss": 0.0866, |
| "num_input_tokens_seen": 190560, |
| "step": 610 |
| }, |
| { |
| "epoch": 6.833333333333333, |
| "grad_norm": 0.0878119021654129, |
| "learning_rate": 1.3866753733777765e-05, |
| "loss": 0.007, |
| "num_input_tokens_seen": 192128, |
| "step": 615 |
| }, |
| { |
| "epoch": 6.888888888888889, |
| "grad_norm": 0.10325942933559418, |
| "learning_rate": 1.343478834970121e-05, |
| "loss": 0.0074, |
| "num_input_tokens_seen": 193664, |
| "step": 620 |
| }, |
| { |
| "epoch": 6.944444444444445, |
| "grad_norm": 2.3258378505706787, |
| "learning_rate": 1.3007172168743854e-05, |
| "loss": 0.0102, |
| "num_input_tokens_seen": 195296, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 0.18902866542339325, |
| "learning_rate": 1.2584065999863102e-05, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 196896, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.1432798057794571, |
| "eval_runtime": 0.5835, |
| "eval_samples_per_second": 68.557, |
| "eval_steps_per_second": 17.139, |
| "num_input_tokens_seen": 196896, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.055555555555555, |
| "grad_norm": 0.10227175801992416, |
| "learning_rate": 1.2165628955985314e-05, |
| "loss": 0.0085, |
| "num_input_tokens_seen": 198528, |
| "step": 635 |
| }, |
| { |
| "epoch": 7.111111111111111, |
| "grad_norm": 0.16049188375473022, |
| "learning_rate": 1.175201839416988e-05, |
| "loss": 0.0203, |
| "num_input_tokens_seen": 200096, |
| "step": 640 |
| }, |
| { |
| "epoch": 7.166666666666667, |
| "grad_norm": 0.05664211884140968, |
| "learning_rate": 1.1343389856433658e-05, |
| "loss": 0.0326, |
| "num_input_tokens_seen": 201696, |
| "step": 645 |
| }, |
| { |
| "epoch": 7.222222222222222, |
| "grad_norm": 0.673611044883728, |
| "learning_rate": 1.0939897011258001e-05, |
| "loss": 0.0038, |
| "num_input_tokens_seen": 203232, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.277777777777778, |
| "grad_norm": 8.291230201721191, |
| "learning_rate": 1.0541691595800337e-05, |
| "loss": 0.084, |
| "num_input_tokens_seen": 204768, |
| "step": 655 |
| }, |
| { |
| "epoch": 7.333333333333333, |
| "grad_norm": 0.0056886388920247555, |
| "learning_rate": 1.0148923358832022e-05, |
| "loss": 0.0079, |
| "num_input_tokens_seen": 206368, |
| "step": 660 |
| }, |
| { |
| "epoch": 7.388888888888889, |
| "grad_norm": 0.005177559796720743, |
| "learning_rate": 9.761740004423927e-06, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 207904, |
| "step": 665 |
| }, |
| { |
| "epoch": 7.444444444444445, |
| "grad_norm": 0.008305308409035206, |
| "learning_rate": 9.380287136401e-06, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 209472, |
| "step": 670 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 0.01929614134132862, |
| "learning_rate": 9.00470820358663e-06, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 211072, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.5, |
| "eval_loss": 0.14391985535621643, |
| "eval_runtime": 0.5862, |
| "eval_samples_per_second": 68.24, |
| "eval_steps_per_second": 17.06, |
| "num_input_tokens_seen": 211072, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.555555555555555, |
| "grad_norm": 0.020041748881340027, |
| "learning_rate": 8.635144445857406e-06, |
| "loss": 0.0137, |
| "num_input_tokens_seen": 212608, |
| "step": 680 |
| }, |
| { |
| "epoch": 7.611111111111111, |
| "grad_norm": 0.05644703283905983, |
| "learning_rate": 8.271734841028553e-06, |
| "loss": 0.0056, |
| "num_input_tokens_seen": 214176, |
| "step": 685 |
| }, |
| { |
| "epoch": 7.666666666666667, |
| "grad_norm": 0.05736091732978821, |
| "learning_rate": 7.914616052590071e-06, |
| "loss": 0.0053, |
| "num_input_tokens_seen": 215744, |
| "step": 690 |
| }, |
| { |
| "epoch": 7.722222222222222, |
| "grad_norm": 1.157248616218567, |
| "learning_rate": 7.563922378313218e-06, |
| "loss": 0.006, |
| "num_input_tokens_seen": 217248, |
| "step": 695 |
| }, |
| { |
| "epoch": 7.777777777777778, |
| "grad_norm": 0.006323433481156826, |
| "learning_rate": 7.219785699746573e-06, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 218816, |
| "step": 700 |
| }, |
| { |
| "epoch": 7.833333333333333, |
| "grad_norm": 0.24250803887844086, |
| "learning_rate": 6.882335432620779e-06, |
| "loss": 0.0738, |
| "num_input_tokens_seen": 220288, |
| "step": 705 |
| }, |
| { |
| "epoch": 7.888888888888889, |
| "grad_norm": 0.04636640101671219, |
| "learning_rate": 6.55169847818059e-06, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 221856, |
| "step": 710 |
| }, |
| { |
| "epoch": 7.944444444444445, |
| "grad_norm": 0.5231844782829285, |
| "learning_rate": 6.22799917546252e-06, |
| "loss": 0.0549, |
| "num_input_tokens_seen": 223424, |
| "step": 715 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 0.016849713400006294, |
| "learning_rate": 5.9113592545359945e-06, |
| "loss": 0.0444, |
| "num_input_tokens_seen": 225056, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.14301875233650208, |
| "eval_runtime": 0.5826, |
| "eval_samples_per_second": 68.663, |
| "eval_steps_per_second": 17.166, |
| "num_input_tokens_seen": 225056, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.055555555555555, |
| "grad_norm": 1.2855383157730103, |
| "learning_rate": 5.601897790725643e-06, |
| "loss": 0.0501, |
| "num_input_tokens_seen": 226656, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.11111111111111, |
| "grad_norm": 0.019593268632888794, |
| "learning_rate": 5.299731159831953e-06, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 228160, |
| "step": 730 |
| }, |
| { |
| "epoch": 8.166666666666666, |
| "grad_norm": 0.03212496265769005, |
| "learning_rate": 5.004972994367102e-06, |
| "loss": 0.0017, |
| "num_input_tokens_seen": 229728, |
| "step": 735 |
| }, |
| { |
| "epoch": 8.222222222222221, |
| "grad_norm": 0.00420254236087203, |
| "learning_rate": 4.7177341408224e-06, |
| "loss": 0.0551, |
| "num_input_tokens_seen": 231328, |
| "step": 740 |
| }, |
| { |
| "epoch": 8.277777777777779, |
| "grad_norm": 0.013452420942485332, |
| "learning_rate": 4.438122617983443e-06, |
| "loss": 0.0728, |
| "num_input_tokens_seen": 232832, |
| "step": 745 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.0068261465057730675, |
| "learning_rate": 4.166243576308712e-06, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 234432, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.38888888888889, |
| "grad_norm": 0.011409962549805641, |
| "learning_rate": 3.9021992583867325e-06, |
| "loss": 0.0058, |
| "num_input_tokens_seen": 235968, |
| "step": 755 |
| }, |
| { |
| "epoch": 8.444444444444445, |
| "grad_norm": 0.07829581946134567, |
| "learning_rate": 3.6460889604868626e-06, |
| "loss": 0.0034, |
| "num_input_tokens_seen": 237568, |
| "step": 760 |
| }, |
| { |
| "epoch": 8.5, |
| "grad_norm": 1.6991177797317505, |
| "learning_rate": 3.398008995217988e-06, |
| "loss": 0.0927, |
| "num_input_tokens_seen": 239168, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.5, |
| "eval_loss": 0.14233675599098206, |
| "eval_runtime": 0.586, |
| "eval_samples_per_second": 68.264, |
| "eval_steps_per_second": 17.066, |
| "num_input_tokens_seen": 239168, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.555555555555555, |
| "grad_norm": 0.19973303377628326, |
| "learning_rate": 3.158052655309332e-06, |
| "loss": 0.0369, |
| "num_input_tokens_seen": 240800, |
| "step": 770 |
| }, |
| { |
| "epoch": 8.61111111111111, |
| "grad_norm": 0.06274840980768204, |
| "learning_rate": 2.9263101785268254e-06, |
| "loss": 0.0027, |
| "num_input_tokens_seen": 242368, |
| "step": 775 |
| }, |
| { |
| "epoch": 8.666666666666666, |
| "grad_norm": 0.12896870076656342, |
| "learning_rate": 2.7028687137384267e-06, |
| "loss": 0.0024, |
| "num_input_tokens_seen": 243872, |
| "step": 780 |
| }, |
| { |
| "epoch": 8.722222222222221, |
| "grad_norm": 0.013180126436054707, |
| "learning_rate": 2.487812288140945e-06, |
| "loss": 0.008, |
| "num_input_tokens_seen": 245472, |
| "step": 785 |
| }, |
| { |
| "epoch": 8.777777777777779, |
| "grad_norm": 0.12642890214920044, |
| "learning_rate": 2.281221775660894e-06, |
| "loss": 0.0008, |
| "num_input_tokens_seen": 247072, |
| "step": 790 |
| }, |
| { |
| "epoch": 8.833333333333334, |
| "grad_norm": 0.011742387898266315, |
| "learning_rate": 2.0831748665410765e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 248640, |
| "step": 795 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "grad_norm": 0.005955726373940706, |
| "learning_rate": 1.893746038124497e-06, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 250208, |
| "step": 800 |
| }, |
| { |
| "epoch": 8.944444444444445, |
| "grad_norm": 0.0940459817647934, |
| "learning_rate": 1.713006526846439e-06, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 251744, |
| "step": 805 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 0.24416488409042358, |
| "learning_rate": 1.541024301445404e-06, |
| "loss": 0.0111, |
| "num_input_tokens_seen": 253312, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.14459559321403503, |
| "eval_runtime": 0.5731, |
| "eval_samples_per_second": 69.794, |
| "eval_steps_per_second": 17.449, |
| "num_input_tokens_seen": 253312, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.055555555555555, |
| "grad_norm": 0.057257551699876785, |
| "learning_rate": 1.3778640374027985e-06, |
| "loss": 0.0048, |
| "num_input_tokens_seen": 254944, |
| "step": 815 |
| }, |
| { |
| "epoch": 9.11111111111111, |
| "grad_norm": 0.006494577042758465, |
| "learning_rate": 1.2235870926211619e-06, |
| "loss": 0.0332, |
| "num_input_tokens_seen": 256512, |
| "step": 820 |
| }, |
| { |
| "epoch": 9.166666666666666, |
| "grad_norm": 0.6156882047653198, |
| "learning_rate": 1.0782514843499653e-06, |
| "loss": 0.0026, |
| "num_input_tokens_seen": 258016, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.222222222222221, |
| "grad_norm": 0.024971961975097656, |
| "learning_rate": 9.419118673676924e-07, |
| "loss": 0.0657, |
| "num_input_tokens_seen": 259552, |
| "step": 830 |
| }, |
| { |
| "epoch": 9.277777777777779, |
| "grad_norm": 0.0062075089663267136, |
| "learning_rate": 8.146195134284052e-07, |
| "loss": 0.0145, |
| "num_input_tokens_seen": 261152, |
| "step": 835 |
| }, |
| { |
| "epoch": 9.333333333333334, |
| "grad_norm": 0.4766158163547516, |
| "learning_rate": 6.964222919805391e-07, |
| "loss": 0.0023, |
| "num_input_tokens_seen": 262720, |
| "step": 840 |
| }, |
| { |
| "epoch": 9.38888888888889, |
| "grad_norm": 0.056195683777332306, |
| "learning_rate": 5.87364652165176e-07, |
| "loss": 0.0087, |
| "num_input_tokens_seen": 264224, |
| "step": 845 |
| }, |
| { |
| "epoch": 9.444444444444445, |
| "grad_norm": 0.9626007080078125, |
| "learning_rate": 4.874876061005173e-07, |
| "loss": 0.0044, |
| "num_input_tokens_seen": 265792, |
| "step": 850 |
| }, |
| { |
| "epoch": 9.5, |
| "grad_norm": 0.15369237959384918, |
| "learning_rate": 3.9682871345891883e-07, |
| "loss": 0.001, |
| "num_input_tokens_seen": 267392, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.5, |
| "eval_loss": 0.1462300568819046, |
| "eval_runtime": 0.5792, |
| "eval_samples_per_second": 69.063, |
| "eval_steps_per_second": 17.266, |
| "num_input_tokens_seen": 267392, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.555555555555555, |
| "grad_norm": 0.028775176033377647, |
| "learning_rate": 3.1542206734221924e-07, |
| "loss": 0.0089, |
| "num_input_tokens_seen": 268896, |
| "step": 860 |
| }, |
| { |
| "epoch": 9.61111111111111, |
| "grad_norm": 0.20606637001037598, |
| "learning_rate": 2.4329828146074095e-07, |
| "loss": 0.0468, |
| "num_input_tokens_seen": 270432, |
| "step": 865 |
| }, |
| { |
| "epoch": 9.666666666666666, |
| "grad_norm": 0.8666467666625977, |
| "learning_rate": 1.8048447862070718e-07, |
| "loss": 0.015, |
| "num_input_tokens_seen": 271968, |
| "step": 870 |
| }, |
| { |
| "epoch": 9.722222222222221, |
| "grad_norm": 0.014927767217159271, |
| "learning_rate": 1.2700428052447033e-07, |
| "loss": 0.0026, |
| "num_input_tokens_seen": 273504, |
| "step": 875 |
| }, |
| { |
| "epoch": 9.777777777777779, |
| "grad_norm": 0.006279713939875364, |
| "learning_rate": 8.28777988873486e-08, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 275104, |
| "step": 880 |
| }, |
| { |
| "epoch": 9.833333333333334, |
| "grad_norm": 1.544179916381836, |
| "learning_rate": 4.8121627874450625e-08, |
| "loss": 0.0058, |
| "num_input_tokens_seen": 276672, |
| "step": 885 |
| }, |
| { |
| "epoch": 9.88888888888889, |
| "grad_norm": 0.714333713054657, |
| "learning_rate": 2.2748837860270267e-08, |
| "loss": 0.0029, |
| "num_input_tokens_seen": 278272, |
| "step": 890 |
| }, |
| { |
| "epoch": 9.944444444444445, |
| "grad_norm": 0.01058835256844759, |
| "learning_rate": 6.768970513457151e-09, |
| "loss": 0.0869, |
| "num_input_tokens_seen": 279808, |
| "step": 895 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.0075642429292202, |
| "learning_rate": 1.8803520859811406e-10, |
| "loss": 0.0256, |
| "num_input_tokens_seen": 281408, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.13989758491516113, |
| "eval_runtime": 0.5834, |
| "eval_samples_per_second": 68.56, |
| "eval_steps_per_second": 17.14, |
| "num_input_tokens_seen": 281408, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 281408, |
| "step": 900, |
| "total_flos": 1.2672124584984576e+16, |
| "train_loss": 0.06401773532231649, |
| "train_runtime": 160.7306, |
| "train_samples_per_second": 22.398, |
| "train_steps_per_second": 5.599 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 900, |
| "num_input_tokens_seen": 281408, |
| "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": 1.2672124584984576e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
