| { |
| "best_global_step": 270, |
| "best_metric": 0.02730442024767399, |
| "best_model_checkpoint": "saves_stability/p-tuning/llama-3-8b-instruct/train_copa_1757340278/checkpoint-270", |
| "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": 71.15585327148438, |
| "learning_rate": 2.2222222222222225e-06, |
| "loss": 3.6939, |
| "num_input_tokens_seen": 1600, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1111111111111111, |
| "grad_norm": 68.3715591430664, |
| "learning_rate": 5e-06, |
| "loss": 2.9524, |
| "num_input_tokens_seen": 3200, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.16666666666666666, |
| "grad_norm": 40.321502685546875, |
| "learning_rate": 7.777777777777777e-06, |
| "loss": 2.1857, |
| "num_input_tokens_seen": 4736, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 45.11865234375, |
| "learning_rate": 1.0555555555555555e-05, |
| "loss": 1.2203, |
| "num_input_tokens_seen": 6368, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.2777777777777778, |
| "grad_norm": 115.53544616699219, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 0.6404, |
| "num_input_tokens_seen": 7872, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 14.631987571716309, |
| "learning_rate": 1.6111111111111115e-05, |
| "loss": 0.419, |
| "num_input_tokens_seen": 9440, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.3888888888888889, |
| "grad_norm": 9.51198673248291, |
| "learning_rate": 1.888888888888889e-05, |
| "loss": 0.2554, |
| "num_input_tokens_seen": 11008, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 23.122678756713867, |
| "learning_rate": 2.1666666666666667e-05, |
| "loss": 0.2464, |
| "num_input_tokens_seen": 12608, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 10.351434707641602, |
| "learning_rate": 2.4444444444444445e-05, |
| "loss": 0.2006, |
| "num_input_tokens_seen": 14144, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 0.1967490166425705, |
| "eval_runtime": 0.9271, |
| "eval_samples_per_second": 43.145, |
| "eval_steps_per_second": 10.786, |
| "num_input_tokens_seen": 14144, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5555555555555556, |
| "grad_norm": 15.461503028869629, |
| "learning_rate": 2.7222222222222223e-05, |
| "loss": 0.2431, |
| "num_input_tokens_seen": 15744, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6111111111111112, |
| "grad_norm": 16.255895614624023, |
| "learning_rate": 3e-05, |
| "loss": 0.6202, |
| "num_input_tokens_seen": 17312, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 2.7334351539611816, |
| "learning_rate": 3.277777777777778e-05, |
| "loss": 0.0652, |
| "num_input_tokens_seen": 18880, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.7222222222222222, |
| "grad_norm": 7.902228832244873, |
| "learning_rate": 3.555555555555556e-05, |
| "loss": 0.0847, |
| "num_input_tokens_seen": 20416, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7777777777777778, |
| "grad_norm": 21.365991592407227, |
| "learning_rate": 3.8333333333333334e-05, |
| "loss": 0.0881, |
| "num_input_tokens_seen": 21920, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 2.1551949977874756, |
| "learning_rate": 4.111111111111111e-05, |
| "loss": 0.0405, |
| "num_input_tokens_seen": 23488, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 0.07108575850725174, |
| "learning_rate": 4.388888888888889e-05, |
| "loss": 0.1006, |
| "num_input_tokens_seen": 25056, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.9444444444444444, |
| "grad_norm": 3.741621732711792, |
| "learning_rate": 4.666666666666667e-05, |
| "loss": 0.2459, |
| "num_input_tokens_seen": 26592, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 15.079221725463867, |
| "learning_rate": 4.9444444444444446e-05, |
| "loss": 0.3225, |
| "num_input_tokens_seen": 28192, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.08561945706605911, |
| "eval_runtime": 0.936, |
| "eval_samples_per_second": 42.737, |
| "eval_steps_per_second": 10.684, |
| "num_input_tokens_seen": 28192, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0555555555555556, |
| "grad_norm": 12.451729774475098, |
| "learning_rate": 4.9996991493233693e-05, |
| "loss": 0.0833, |
| "num_input_tokens_seen": 29792, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 6.963460445404053, |
| "learning_rate": 4.99847706754774e-05, |
| "loss": 0.0913, |
| "num_input_tokens_seen": 31360, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.1666666666666667, |
| "grad_norm": 0.018313022330403328, |
| "learning_rate": 4.9963154107272295e-05, |
| "loss": 0.0017, |
| "num_input_tokens_seen": 32832, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.2222222222222223, |
| "grad_norm": 0.010015306994318962, |
| "learning_rate": 4.993214991772563e-05, |
| "loss": 0.108, |
| "num_input_tokens_seen": 34336, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.2777777777777777, |
| "grad_norm": 1.3597711324691772, |
| "learning_rate": 4.989176976624511e-05, |
| "loss": 0.0895, |
| "num_input_tokens_seen": 35936, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 0.020921070128679276, |
| "learning_rate": 4.9842028838154285e-05, |
| "loss": 0.0142, |
| "num_input_tokens_seen": 37536, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.3888888888888888, |
| "grad_norm": 0.05943799018859863, |
| "learning_rate": 4.978294583898196e-05, |
| "loss": 0.0031, |
| "num_input_tokens_seen": 39040, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.4444444444444444, |
| "grad_norm": 194.86752319335938, |
| "learning_rate": 4.971454298742779e-05, |
| "loss": 0.7974, |
| "num_input_tokens_seen": 40576, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 0.8524989485740662, |
| "learning_rate": 4.963684600700679e-05, |
| "loss": 0.4327, |
| "num_input_tokens_seen": 42208, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5, |
| "eval_loss": 0.047824930399656296, |
| "eval_runtime": 0.9383, |
| "eval_samples_per_second": 42.632, |
| "eval_steps_per_second": 10.658, |
| "num_input_tokens_seen": 42208, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 12.642061233520508, |
| "learning_rate": 4.9549884116375714e-05, |
| "loss": 0.2017, |
| "num_input_tokens_seen": 43776, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.6111111111111112, |
| "grad_norm": 0.026874296367168427, |
| "learning_rate": 4.9453690018345144e-05, |
| "loss": 0.0376, |
| "num_input_tokens_seen": 45376, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 0.2205730527639389, |
| "learning_rate": 4.934829988758131e-05, |
| "loss": 0.2751, |
| "num_input_tokens_seen": 46944, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.7222222222222223, |
| "grad_norm": 4.115603923797607, |
| "learning_rate": 4.923375335700223e-05, |
| "loss": 0.1317, |
| "num_input_tokens_seen": 48512, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 0.01587647944688797, |
| "learning_rate": 4.9110093502873476e-05, |
| "loss": 0.0702, |
| "num_input_tokens_seen": 50112, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.8333333333333335, |
| "grad_norm": 8.227548599243164, |
| "learning_rate": 4.897736682860885e-05, |
| "loss": 0.2497, |
| "num_input_tokens_seen": 51680, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.8888888888888888, |
| "grad_norm": 0.40967798233032227, |
| "learning_rate": 4.883562324728241e-05, |
| "loss": 0.0336, |
| "num_input_tokens_seen": 53216, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.9444444444444444, |
| "grad_norm": 1.227698802947998, |
| "learning_rate": 4.868491606285823e-05, |
| "loss": 0.1445, |
| "num_input_tokens_seen": 54720, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 3.000871419906616, |
| "learning_rate": 4.8525301950144894e-05, |
| "loss": 0.0202, |
| "num_input_tokens_seen": 56256, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.07754607498645782, |
| "eval_runtime": 0.9351, |
| "eval_samples_per_second": 42.775, |
| "eval_steps_per_second": 10.694, |
| "num_input_tokens_seen": 56256, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0555555555555554, |
| "grad_norm": 0.086800716817379, |
| "learning_rate": 4.835684093348244e-05, |
| "loss": 0.0017, |
| "num_input_tokens_seen": 57856, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.111111111111111, |
| "grad_norm": 2.6525955200195312, |
| "learning_rate": 4.817959636416969e-05, |
| "loss": 0.0282, |
| "num_input_tokens_seen": 59456, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.1666666666666665, |
| "grad_norm": 4.272388458251953, |
| "learning_rate": 4.7993634896640394e-05, |
| "loss": 0.0678, |
| "num_input_tokens_seen": 61024, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 0.2532036304473877, |
| "learning_rate": 4.779902646339722e-05, |
| "loss": 0.014, |
| "num_input_tokens_seen": 62592, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.2777777777777777, |
| "grad_norm": 16.97927474975586, |
| "learning_rate": 4.759584424871302e-05, |
| "loss": 0.0197, |
| "num_input_tokens_seen": 64128, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 0.13003849983215332, |
| "learning_rate": 4.7384164661109176e-05, |
| "loss": 0.131, |
| "num_input_tokens_seen": 65696, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.388888888888889, |
| "grad_norm": 0.07962575554847717, |
| "learning_rate": 4.7164067304621536e-05, |
| "loss": 0.1118, |
| "num_input_tokens_seen": 67200, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 0.15706808865070343, |
| "learning_rate": 4.693563494886455e-05, |
| "loss": 0.0138, |
| "num_input_tokens_seen": 68800, |
| "step": 220 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 0.27382174134254456, |
| "learning_rate": 4.669895349790502e-05, |
| "loss": 0.1742, |
| "num_input_tokens_seen": 70368, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.05516480281949043, |
| "eval_runtime": 0.9368, |
| "eval_samples_per_second": 42.699, |
| "eval_steps_per_second": 10.675, |
| "num_input_tokens_seen": 70368, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5555555555555554, |
| "grad_norm": 5.053715229034424, |
| "learning_rate": 4.645411195795709e-05, |
| "loss": 0.0405, |
| "num_input_tokens_seen": 71872, |
| "step": 230 |
| }, |
| { |
| "epoch": 2.611111111111111, |
| "grad_norm": 0.36858999729156494, |
| "learning_rate": 4.620120240391065e-05, |
| "loss": 0.0023, |
| "num_input_tokens_seen": 73472, |
| "step": 235 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 7.379425048828125, |
| "learning_rate": 4.5940319944705736e-05, |
| "loss": 0.0424, |
| "num_input_tokens_seen": 75040, |
| "step": 240 |
| }, |
| { |
| "epoch": 2.7222222222222223, |
| "grad_norm": 17.919721603393555, |
| "learning_rate": 4.567156268756594e-05, |
| "loss": 0.023, |
| "num_input_tokens_seen": 76544, |
| "step": 245 |
| }, |
| { |
| "epoch": 2.7777777777777777, |
| "grad_norm": 0.3826093375682831, |
| "learning_rate": 4.539503170110431e-05, |
| "loss": 0.0064, |
| "num_input_tokens_seen": 78048, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.8333333333333335, |
| "grad_norm": 0.002257711486890912, |
| "learning_rate": 4.5110830977315556e-05, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 79680, |
| "step": 255 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 0.008002882823348045, |
| "learning_rate": 4.4819067392468944e-05, |
| "loss": 0.051, |
| "num_input_tokens_seen": 81184, |
| "step": 260 |
| }, |
| { |
| "epoch": 2.9444444444444446, |
| "grad_norm": 4.711673736572266, |
| "learning_rate": 4.4519850666916484e-05, |
| "loss": 0.1127, |
| "num_input_tokens_seen": 82784, |
| "step": 265 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.012198343873023987, |
| "learning_rate": 4.4213293323831585e-05, |
| "loss": 0.0049, |
| "num_input_tokens_seen": 84320, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.02730442024767399, |
| "eval_runtime": 0.9347, |
| "eval_samples_per_second": 42.793, |
| "eval_steps_per_second": 10.698, |
| "num_input_tokens_seen": 84320, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0555555555555554, |
| "grad_norm": 0.0810229554772377, |
| "learning_rate": 4.38995106468937e-05, |
| "loss": 0.0102, |
| "num_input_tokens_seen": 85920, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 0.026958152651786804, |
| "learning_rate": 4.357862063693486e-05, |
| "loss": 0.0016, |
| "num_input_tokens_seen": 87456, |
| "step": 280 |
| }, |
| { |
| "epoch": 3.1666666666666665, |
| "grad_norm": 0.0024806878063827753, |
| "learning_rate": 4.325074396756437e-05, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 89056, |
| "step": 285 |
| }, |
| { |
| "epoch": 3.2222222222222223, |
| "grad_norm": 0.04719718173146248, |
| "learning_rate": 4.2916003939788403e-05, |
| "loss": 0.0083, |
| "num_input_tokens_seen": 90560, |
| "step": 290 |
| }, |
| { |
| "epoch": 3.2777777777777777, |
| "grad_norm": 3.299919605255127, |
| "learning_rate": 4.257452643564155e-05, |
| "loss": 0.0067, |
| "num_input_tokens_seen": 92128, |
| "step": 295 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.016195300966501236, |
| "learning_rate": 4.22264398708477e-05, |
| "loss": 0.0141, |
| "num_input_tokens_seen": 93728, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.388888888888889, |
| "grad_norm": 31.858551025390625, |
| "learning_rate": 4.1871875146528195e-05, |
| "loss": 0.0285, |
| "num_input_tokens_seen": 95264, |
| "step": 305 |
| }, |
| { |
| "epoch": 3.4444444444444446, |
| "grad_norm": 0.014500042423605919, |
| "learning_rate": 4.1510965599975196e-05, |
| "loss": 0.1069, |
| "num_input_tokens_seen": 96768, |
| "step": 310 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 0.30705544352531433, |
| "learning_rate": 4.114384695450906e-05, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 98400, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5, |
| "eval_loss": 0.058323830366134644, |
| "eval_runtime": 0.9375, |
| "eval_samples_per_second": 42.668, |
| "eval_steps_per_second": 10.667, |
| "num_input_tokens_seen": 98400, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 19.586668014526367, |
| "learning_rate": 4.077065726843828e-05, |
| "loss": 0.0466, |
| "num_input_tokens_seen": 99968, |
| "step": 320 |
| }, |
| { |
| "epoch": 3.611111111111111, |
| "grad_norm": 1.009904146194458, |
| "learning_rate": 4.039153688314145e-05, |
| "loss": 0.0043, |
| "num_input_tokens_seen": 101504, |
| "step": 325 |
| }, |
| { |
| "epoch": 3.6666666666666665, |
| "grad_norm": 0.2828821539878845, |
| "learning_rate": 4.000662837029062e-05, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 103008, |
| "step": 330 |
| }, |
| { |
| "epoch": 3.7222222222222223, |
| "grad_norm": 0.0027755468618124723, |
| "learning_rate": 3.961607647823583e-05, |
| "loss": 0.0087, |
| "num_input_tokens_seen": 104608, |
| "step": 335 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 0.1006738692522049, |
| "learning_rate": 3.9220028077571295e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 106144, |
| "step": 340 |
| }, |
| { |
| "epoch": 3.8333333333333335, |
| "grad_norm": 0.005183361936360598, |
| "learning_rate": 3.881863210590332e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 107744, |
| "step": 345 |
| }, |
| { |
| "epoch": 3.888888888888889, |
| "grad_norm": 13.548725128173828, |
| "learning_rate": 3.841203951184095e-05, |
| "loss": 0.1647, |
| "num_input_tokens_seen": 109280, |
| "step": 350 |
| }, |
| { |
| "epoch": 3.9444444444444446, |
| "grad_norm": 0.006770310923457146, |
| "learning_rate": 3.8000403198230387e-05, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 110848, |
| "step": 355 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.02527625858783722, |
| "learning_rate": 3.75838779646545e-05, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 112416, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.03324687108397484, |
| "eval_runtime": 0.936, |
| "eval_samples_per_second": 42.735, |
| "eval_steps_per_second": 10.684, |
| "num_input_tokens_seen": 112416, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.055555555555555, |
| "grad_norm": 0.7226027846336365, |
| "learning_rate": 3.7162620449219e-05, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 113984, |
| "step": 365 |
| }, |
| { |
| "epoch": 4.111111111111111, |
| "grad_norm": 0.0285462886095047, |
| "learning_rate": 3.673678906964727e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 115552, |
| "step": 370 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 0.004855355713516474, |
| "learning_rate": 3.630654396370594e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 117152, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 0.027779415249824524, |
| "learning_rate": 3.5872046928983626e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 118784, |
| "step": 380 |
| }, |
| { |
| "epoch": 4.277777777777778, |
| "grad_norm": 0.015533392317593098, |
| "learning_rate": 3.543346136204545e-05, |
| "loss": 0.0433, |
| "num_input_tokens_seen": 120256, |
| "step": 385 |
| }, |
| { |
| "epoch": 4.333333333333333, |
| "grad_norm": 0.005797127727419138, |
| "learning_rate": 3.499095219698631e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 121824, |
| "step": 390 |
| }, |
| { |
| "epoch": 4.388888888888889, |
| "grad_norm": 0.0747162252664566, |
| "learning_rate": 3.454468584340588e-05, |
| "loss": 0.0026, |
| "num_input_tokens_seen": 123392, |
| "step": 395 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 0.05891333147883415, |
| "learning_rate": 3.409483012382879e-05, |
| "loss": 0.0398, |
| "num_input_tokens_seen": 124928, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 0.027306361123919487, |
| "learning_rate": 3.364155421059342e-05, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 126496, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.5, |
| "eval_loss": 0.04061094671487808, |
| "eval_runtime": 0.9384, |
| "eval_samples_per_second": 42.627, |
| "eval_steps_per_second": 10.657, |
| "num_input_tokens_seen": 126496, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.555555555555555, |
| "grad_norm": 0.0021955284755676985, |
| "learning_rate": 3.318502856223311e-05, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 128064, |
| "step": 410 |
| }, |
| { |
| "epoch": 4.611111111111111, |
| "grad_norm": 0.02812635712325573, |
| "learning_rate": 3.272542485937369e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 129632, |
| "step": 415 |
| }, |
| { |
| "epoch": 4.666666666666667, |
| "grad_norm": 0.022411026060581207, |
| "learning_rate": 3.2262915940171376e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 131200, |
| "step": 420 |
| }, |
| { |
| "epoch": 4.722222222222222, |
| "grad_norm": 0.001348581979982555, |
| "learning_rate": 3.1797675735315455e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 132800, |
| "step": 425 |
| }, |
| { |
| "epoch": 4.777777777777778, |
| "grad_norm": 0.004968563094735146, |
| "learning_rate": 3.132987920262005e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 134336, |
| "step": 430 |
| }, |
| { |
| "epoch": 4.833333333333333, |
| "grad_norm": 0.009831856936216354, |
| "learning_rate": 3.085970226122962e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 135904, |
| "step": 435 |
| }, |
| { |
| "epoch": 4.888888888888889, |
| "grad_norm": 0.0047363354824483395, |
| "learning_rate": 3.0387321725463e-05, |
| "loss": 0.0177, |
| "num_input_tokens_seen": 137504, |
| "step": 440 |
| }, |
| { |
| "epoch": 4.944444444444445, |
| "grad_norm": 0.010216707363724709, |
| "learning_rate": 2.9912915238320754e-05, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 139040, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.011602720245718956, |
| "learning_rate": 2.9436661204680882e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 140544, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.03640992194414139, |
| "eval_runtime": 0.9326, |
| "eval_samples_per_second": 42.889, |
| "eval_steps_per_second": 10.722, |
| "num_input_tokens_seen": 140544, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.055555555555555, |
| "grad_norm": 0.04442625865340233, |
| "learning_rate": 2.8958738724208072e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 142176, |
| "step": 455 |
| }, |
| { |
| "epoch": 5.111111111111111, |
| "grad_norm": 0.0018977177096530795, |
| "learning_rate": 2.8479327524001636e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 143776, |
| "step": 460 |
| }, |
| { |
| "epoch": 5.166666666666667, |
| "grad_norm": 0.05888612940907478, |
| "learning_rate": 2.7998607891007495e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 145312, |
| "step": 465 |
| }, |
| { |
| "epoch": 5.222222222222222, |
| "grad_norm": 0.008200118318200111, |
| "learning_rate": 2.7516760604219617e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 146880, |
| "step": 470 |
| }, |
| { |
| "epoch": 5.277777777777778, |
| "grad_norm": 0.17480984330177307, |
| "learning_rate": 2.7033966866696457e-05, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 148416, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "grad_norm": 0.007679934147745371, |
| "learning_rate": 2.6550408237417885e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 149952, |
| "step": 480 |
| }, |
| { |
| "epoch": 5.388888888888889, |
| "grad_norm": 0.007722547277808189, |
| "learning_rate": 2.6066266563008267e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 151488, |
| "step": 485 |
| }, |
| { |
| "epoch": 5.444444444444445, |
| "grad_norm": 0.002113697351887822, |
| "learning_rate": 2.5581723909351406e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 153024, |
| "step": 490 |
| }, |
| { |
| "epoch": 5.5, |
| "grad_norm": 0.0018104183254763484, |
| "learning_rate": 2.5096962493123012e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 154592, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.5, |
| "eval_loss": 0.047297000885009766, |
| "eval_runtime": 0.9407, |
| "eval_samples_per_second": 42.522, |
| "eval_steps_per_second": 10.63, |
| "num_input_tokens_seen": 154592, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.555555555555555, |
| "grad_norm": 0.014669693075120449, |
| "learning_rate": 2.461216461326642e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 156128, |
| "step": 500 |
| }, |
| { |
| "epoch": 5.611111111111111, |
| "grad_norm": 0.1601002961397171, |
| "learning_rate": 2.4127512582437485e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 157728, |
| "step": 505 |
| }, |
| { |
| "epoch": 5.666666666666667, |
| "grad_norm": 0.04238174855709076, |
| "learning_rate": 2.364318865844416e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 159296, |
| "step": 510 |
| }, |
| { |
| "epoch": 5.722222222222222, |
| "grad_norm": 0.002438083989545703, |
| "learning_rate": 2.3159374975706884e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 160896, |
| "step": 515 |
| }, |
| { |
| "epoch": 5.777777777777778, |
| "grad_norm": 0.008906746283173561, |
| "learning_rate": 2.2676253476765196e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 162400, |
| "step": 520 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 0.0016734261298552155, |
| "learning_rate": 2.2194005843856636e-05, |
| "loss": 0.0319, |
| "num_input_tokens_seen": 163968, |
| "step": 525 |
| }, |
| { |
| "epoch": 5.888888888888889, |
| "grad_norm": 0.12285536527633667, |
| "learning_rate": 2.1712813430593436e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 165536, |
| "step": 530 |
| }, |
| { |
| "epoch": 5.944444444444445, |
| "grad_norm": 0.003246977925300598, |
| "learning_rate": 2.1232857193762924e-05, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 167168, |
| "step": 535 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.005042417906224728, |
| "learning_rate": 2.0754317625276983e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 168768, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.04458264261484146, |
| "eval_runtime": 0.935, |
| "eval_samples_per_second": 42.779, |
| "eval_steps_per_second": 10.695, |
| "num_input_tokens_seen": 168768, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.055555555555555, |
| "grad_norm": 0.0037654147017747164, |
| "learning_rate": 2.02773746842965e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 170336, |
| "step": 545 |
| }, |
| { |
| "epoch": 6.111111111111111, |
| "grad_norm": 0.0034227280411869287, |
| "learning_rate": 1.980220772955602e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 171904, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.166666666666667, |
| "grad_norm": 0.0037588723935186863, |
| "learning_rate": 1.932899545191433e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 173440, |
| "step": 555 |
| }, |
| { |
| "epoch": 6.222222222222222, |
| "grad_norm": 0.0026463493704795837, |
| "learning_rate": 1.8857915807156092e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 174976, |
| "step": 560 |
| }, |
| { |
| "epoch": 6.277777777777778, |
| "grad_norm": 0.0019206305732950568, |
| "learning_rate": 1.838914594906995e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 176608, |
| "step": 565 |
| }, |
| { |
| "epoch": 6.333333333333333, |
| "grad_norm": 0.0014090255135670304, |
| "learning_rate": 1.792286216282824e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 178144, |
| "step": 570 |
| }, |
| { |
| "epoch": 6.388888888888889, |
| "grad_norm": 0.002481313655152917, |
| "learning_rate": 1.7459239798693364e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 179680, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.444444444444445, |
| "grad_norm": 0.008465290069580078, |
| "learning_rate": 1.699845320607571e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 181248, |
| "step": 580 |
| }, |
| { |
| "epoch": 6.5, |
| "grad_norm": 0.001316405599936843, |
| "learning_rate": 1.6540675667967974e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 182848, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.5, |
| "eval_loss": 0.04233887046575546, |
| "eval_runtime": 0.9401, |
| "eval_samples_per_second": 42.55, |
| "eval_steps_per_second": 10.637, |
| "num_input_tokens_seen": 182848, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.555555555555555, |
| "grad_norm": 0.04860655218362808, |
| "learning_rate": 1.60860793357805e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 184352, |
| "step": 590 |
| }, |
| { |
| "epoch": 6.611111111111111, |
| "grad_norm": 0.0009187149698846042, |
| "learning_rate": 1.56348351646022e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 185952, |
| "step": 595 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.002530870260670781, |
| "learning_rate": 1.5187112848911323e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 187488, |
| "step": 600 |
| }, |
| { |
| "epoch": 6.722222222222222, |
| "grad_norm": 0.0012527881190180779, |
| "learning_rate": 1.47430807587603e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 189056, |
| "step": 605 |
| }, |
| { |
| "epoch": 6.777777777777778, |
| "grad_norm": 0.001099999062716961, |
| "learning_rate": 1.430290587645865e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 190592, |
| "step": 610 |
| }, |
| { |
| "epoch": 6.833333333333333, |
| "grad_norm": 0.0013586750719696283, |
| "learning_rate": 1.3866753733777765e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 192192, |
| "step": 615 |
| }, |
| { |
| "epoch": 6.888888888888889, |
| "grad_norm": 0.0013695342931896448, |
| "learning_rate": 1.343478834970121e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 193824, |
| "step": 620 |
| }, |
| { |
| "epoch": 6.944444444444445, |
| "grad_norm": 0.0010807205690070987, |
| "learning_rate": 1.3007172168743854e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 195392, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 0.0012982726329937577, |
| "learning_rate": 1.2584065999863102e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 196896, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.04652819782495499, |
| "eval_runtime": 0.9398, |
| "eval_samples_per_second": 42.56, |
| "eval_steps_per_second": 10.64, |
| "num_input_tokens_seen": 196896, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.055555555555555, |
| "grad_norm": 0.001132346922531724, |
| "learning_rate": 1.2165628955985314e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 198496, |
| "step": 635 |
| }, |
| { |
| "epoch": 7.111111111111111, |
| "grad_norm": 0.0008002260001376271, |
| "learning_rate": 1.175201839416988e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 200064, |
| "step": 640 |
| }, |
| { |
| "epoch": 7.166666666666667, |
| "grad_norm": 0.0009282976971007884, |
| "learning_rate": 1.1343389856433658e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 201600, |
| "step": 645 |
| }, |
| { |
| "epoch": 7.222222222222222, |
| "grad_norm": 0.0008463115664198995, |
| "learning_rate": 1.0939897011258001e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 203168, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.277777777777778, |
| "grad_norm": 0.007376443594694138, |
| "learning_rate": 1.0541691595800337e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 204704, |
| "step": 655 |
| }, |
| { |
| "epoch": 7.333333333333333, |
| "grad_norm": 0.0009704413823783398, |
| "learning_rate": 1.0148923358832022e-05, |
| "loss": 0.0, |
| "num_input_tokens_seen": 206272, |
| "step": 660 |
| }, |
| { |
| "epoch": 7.388888888888889, |
| "grad_norm": 0.0008887411095201969, |
| "learning_rate": 9.761740004423927e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 207872, |
| "step": 665 |
| }, |
| { |
| "epoch": 7.444444444444445, |
| "grad_norm": 0.0008462558034807444, |
| "learning_rate": 9.380287136401e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 209408, |
| "step": 670 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 0.0009714352781884372, |
| "learning_rate": 9.00470820358663e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 210912, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.5, |
| "eval_loss": 0.04352349787950516, |
| "eval_runtime": 0.9404, |
| "eval_samples_per_second": 42.537, |
| "eval_steps_per_second": 10.634, |
| "num_input_tokens_seen": 210912, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.555555555555555, |
| "grad_norm": 0.022096682339906693, |
| "learning_rate": 8.635144445857406e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 212480, |
| "step": 680 |
| }, |
| { |
| "epoch": 7.611111111111111, |
| "grad_norm": 0.0007414401625283062, |
| "learning_rate": 8.271734841028553e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 214016, |
| "step": 685 |
| }, |
| { |
| "epoch": 7.666666666666667, |
| "grad_norm": 0.0009747538715600967, |
| "learning_rate": 7.914616052590071e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 215584, |
| "step": 690 |
| }, |
| { |
| "epoch": 7.722222222222222, |
| "grad_norm": 0.0014644165057688951, |
| "learning_rate": 7.563922378313218e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 217216, |
| "step": 695 |
| }, |
| { |
| "epoch": 7.777777777777778, |
| "grad_norm": 0.014968394301831722, |
| "learning_rate": 7.219785699746573e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 218752, |
| "step": 700 |
| }, |
| { |
| "epoch": 7.833333333333333, |
| "grad_norm": 0.0009418909321539104, |
| "learning_rate": 6.882335432620779e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 220256, |
| "step": 705 |
| }, |
| { |
| "epoch": 7.888888888888889, |
| "grad_norm": 0.0038506044074892998, |
| "learning_rate": 6.55169847818059e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 221792, |
| "step": 710 |
| }, |
| { |
| "epoch": 7.944444444444445, |
| "grad_norm": 0.001018329057842493, |
| "learning_rate": 6.22799917546252e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 223424, |
| "step": 715 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 0.0008915183134377003, |
| "learning_rate": 5.9113592545359945e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 225024, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.04278353974223137, |
| "eval_runtime": 0.9373, |
| "eval_samples_per_second": 42.675, |
| "eval_steps_per_second": 10.669, |
| "num_input_tokens_seen": 225024, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.055555555555555, |
| "grad_norm": 0.0008386970730498433, |
| "learning_rate": 5.601897790725643e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 226560, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.11111111111111, |
| "grad_norm": 0.0008619049331173301, |
| "learning_rate": 5.299731159831953e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 228128, |
| "step": 730 |
| }, |
| { |
| "epoch": 8.166666666666666, |
| "grad_norm": 0.0008686182554811239, |
| "learning_rate": 5.004972994367102e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 229728, |
| "step": 735 |
| }, |
| { |
| "epoch": 8.222222222222221, |
| "grad_norm": 0.0009698973735794425, |
| "learning_rate": 4.7177341408224e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 231328, |
| "step": 740 |
| }, |
| { |
| "epoch": 8.277777777777779, |
| "grad_norm": 0.0007991605089046061, |
| "learning_rate": 4.438122617983443e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 232896, |
| "step": 745 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.0008160381694324315, |
| "learning_rate": 4.166243576308712e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 234496, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.38888888888889, |
| "grad_norm": 0.0007699804846197367, |
| "learning_rate": 3.9021992583867325e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 236064, |
| "step": 755 |
| }, |
| { |
| "epoch": 8.444444444444445, |
| "grad_norm": 0.005575160961598158, |
| "learning_rate": 3.6460889604868626e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 237632, |
| "step": 760 |
| }, |
| { |
| "epoch": 8.5, |
| "grad_norm": 0.01922343671321869, |
| "learning_rate": 3.398008995217988e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 239200, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.5, |
| "eval_loss": 0.045287247747182846, |
| "eval_runtime": 0.9348, |
| "eval_samples_per_second": 42.789, |
| "eval_steps_per_second": 10.697, |
| "num_input_tokens_seen": 239200, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.555555555555555, |
| "grad_norm": 0.0008846694836392999, |
| "learning_rate": 3.158052655309332e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 240832, |
| "step": 770 |
| }, |
| { |
| "epoch": 8.61111111111111, |
| "grad_norm": 0.0021270590368658304, |
| "learning_rate": 2.9263101785268254e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 242304, |
| "step": 775 |
| }, |
| { |
| "epoch": 8.666666666666666, |
| "grad_norm": 0.0008760837372392416, |
| "learning_rate": 2.7028687137384267e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 243840, |
| "step": 780 |
| }, |
| { |
| "epoch": 8.722222222222221, |
| "grad_norm": 0.0007243354339152575, |
| "learning_rate": 2.487812288140945e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 245376, |
| "step": 785 |
| }, |
| { |
| "epoch": 8.777777777777779, |
| "grad_norm": 0.0008720310870558023, |
| "learning_rate": 2.281221775660894e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 246880, |
| "step": 790 |
| }, |
| { |
| "epoch": 8.833333333333334, |
| "grad_norm": 0.0007376865250989795, |
| "learning_rate": 2.0831748665410765e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 248448, |
| "step": 795 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "grad_norm": 0.0008655358105897903, |
| "learning_rate": 1.893746038124497e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 250016, |
| "step": 800 |
| }, |
| { |
| "epoch": 8.944444444444445, |
| "grad_norm": 0.0012015948304906487, |
| "learning_rate": 1.713006526846439e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 251584, |
| "step": 805 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 0.0009060095762833953, |
| "learning_rate": 1.541024301445404e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 253152, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.044300518929958344, |
| "eval_runtime": 0.9422, |
| "eval_samples_per_second": 42.455, |
| "eval_steps_per_second": 10.614, |
| "num_input_tokens_seen": 253152, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.055555555555555, |
| "grad_norm": 0.0007921405485831201, |
| "learning_rate": 1.3778640374027985e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 254720, |
| "step": 815 |
| }, |
| { |
| "epoch": 9.11111111111111, |
| "grad_norm": 0.013191518373787403, |
| "learning_rate": 1.2235870926211619e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 256288, |
| "step": 820 |
| }, |
| { |
| "epoch": 9.166666666666666, |
| "grad_norm": 0.0007800323655828834, |
| "learning_rate": 1.0782514843499653e-06, |
| "loss": 0.0, |
| "num_input_tokens_seen": 257824, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.222222222222221, |
| "grad_norm": 0.0007931567379273474, |
| "learning_rate": 9.419118673676924e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 259392, |
| "step": 830 |
| }, |
| { |
| "epoch": 9.277777777777779, |
| "grad_norm": 0.010378471575677395, |
| "learning_rate": 8.146195134284052e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 260896, |
| "step": 835 |
| }, |
| { |
| "epoch": 9.333333333333334, |
| "grad_norm": 0.0008987840847112238, |
| "learning_rate": 6.964222919805391e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 262464, |
| "step": 840 |
| }, |
| { |
| "epoch": 9.38888888888889, |
| "grad_norm": 0.000674614158924669, |
| "learning_rate": 5.87364652165176e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 264000, |
| "step": 845 |
| }, |
| { |
| "epoch": 9.444444444444445, |
| "grad_norm": 0.010154387913644314, |
| "learning_rate": 4.874876061005173e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 265536, |
| "step": 850 |
| }, |
| { |
| "epoch": 9.5, |
| "grad_norm": 0.0008763263467699289, |
| "learning_rate": 3.9682871345891883e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 267040, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.5, |
| "eval_loss": 0.04950767755508423, |
| "eval_runtime": 0.9432, |
| "eval_samples_per_second": 42.407, |
| "eval_steps_per_second": 10.602, |
| "num_input_tokens_seen": 267040, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.555555555555555, |
| "grad_norm": 0.0009065108606591821, |
| "learning_rate": 3.1542206734221924e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 268576, |
| "step": 860 |
| }, |
| { |
| "epoch": 9.61111111111111, |
| "grad_norm": 0.000811050646007061, |
| "learning_rate": 2.4329828146074095e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 270144, |
| "step": 865 |
| }, |
| { |
| "epoch": 9.666666666666666, |
| "grad_norm": 0.0011837816564366221, |
| "learning_rate": 1.8048447862070718e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 271648, |
| "step": 870 |
| }, |
| { |
| "epoch": 9.722222222222221, |
| "grad_norm": 0.0008325163507834077, |
| "learning_rate": 1.2700428052447033e-07, |
| "loss": 0.0, |
| "num_input_tokens_seen": 273280, |
| "step": 875 |
| }, |
| { |
| "epoch": 9.777777777777779, |
| "grad_norm": 0.024414081126451492, |
| "learning_rate": 8.28777988873486e-08, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 274880, |
| "step": 880 |
| }, |
| { |
| "epoch": 9.833333333333334, |
| "grad_norm": 0.0009835021337494254, |
| "learning_rate": 4.8121627874450625e-08, |
| "loss": 0.0, |
| "num_input_tokens_seen": 276480, |
| "step": 885 |
| }, |
| { |
| "epoch": 9.88888888888889, |
| "grad_norm": 0.0007079241331666708, |
| "learning_rate": 2.2748837860270267e-08, |
| "loss": 0.0, |
| "num_input_tokens_seen": 278080, |
| "step": 890 |
| }, |
| { |
| "epoch": 9.944444444444445, |
| "grad_norm": 0.0008622616878710687, |
| "learning_rate": 6.768970513457151e-09, |
| "loss": 0.0, |
| "num_input_tokens_seen": 279680, |
| "step": 895 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.014377212151885033, |
| "learning_rate": 1.8803520859811406e-10, |
| "loss": 0.0, |
| "num_input_tokens_seen": 281312, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.04835839569568634, |
| "eval_runtime": 0.9356, |
| "eval_samples_per_second": 42.755, |
| "eval_steps_per_second": 10.689, |
| "num_input_tokens_seen": 281312, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 281312, |
| "step": 900, |
| "total_flos": 1.2667352215977984e+16, |
| "train_loss": 0.09896362295126084, |
| "train_runtime": 205.0036, |
| "train_samples_per_second": 17.561, |
| "train_steps_per_second": 4.39 |
| } |
| ], |
| "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 |
| } |
|
|
