| { |
| "best_global_step": 90, |
| "best_metric": 0.11327455192804337, |
| "best_model_checkpoint": "saves_stability/prompt-tuning/llama-3-8b-instruct/train_copa_1757340228/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": 17.75, |
| "learning_rate": 2.2222222222222225e-06, |
| "loss": 0.3036, |
| "num_input_tokens_seen": 1568, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1111111111111111, |
| "grad_norm": 22.5, |
| "learning_rate": 5e-06, |
| "loss": 0.4605, |
| "num_input_tokens_seen": 3104, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.16666666666666666, |
| "grad_norm": 16.25, |
| "learning_rate": 7.777777777777777e-06, |
| "loss": 0.6002, |
| "num_input_tokens_seen": 4704, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 6.0, |
| "learning_rate": 1.0555555555555555e-05, |
| "loss": 0.284, |
| "num_input_tokens_seen": 6304, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.2777777777777778, |
| "grad_norm": 18.0, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 0.3611, |
| "num_input_tokens_seen": 7808, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.6111111111111115e-05, |
| "loss": 0.218, |
| "num_input_tokens_seen": 9344, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.3888888888888889, |
| "grad_norm": 2.34375, |
| "learning_rate": 1.888888888888889e-05, |
| "loss": 0.2672, |
| "num_input_tokens_seen": 10912, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 1.96875, |
| "learning_rate": 2.1666666666666667e-05, |
| "loss": 0.1508, |
| "num_input_tokens_seen": 12512, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 4.6875, |
| "learning_rate": 2.4444444444444445e-05, |
| "loss": 0.1925, |
| "num_input_tokens_seen": 14016, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 0.14511804282665253, |
| "eval_runtime": 0.8578, |
| "eval_samples_per_second": 46.633, |
| "eval_steps_per_second": 11.658, |
| "num_input_tokens_seen": 14016, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5555555555555556, |
| "grad_norm": 8.375, |
| "learning_rate": 2.7222222222222223e-05, |
| "loss": 0.0208, |
| "num_input_tokens_seen": 15552, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6111111111111112, |
| "grad_norm": 1.2578125, |
| "learning_rate": 3e-05, |
| "loss": 0.0363, |
| "num_input_tokens_seen": 17120, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 9.25, |
| "learning_rate": 3.277777777777778e-05, |
| "loss": 0.1339, |
| "num_input_tokens_seen": 18720, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.7222222222222222, |
| "grad_norm": 2.703125, |
| "learning_rate": 3.555555555555556e-05, |
| "loss": 0.1454, |
| "num_input_tokens_seen": 20288, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7777777777777778, |
| "grad_norm": 3.234375, |
| "learning_rate": 3.8333333333333334e-05, |
| "loss": 0.0142, |
| "num_input_tokens_seen": 21856, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 20.375, |
| "learning_rate": 4.111111111111111e-05, |
| "loss": 0.1892, |
| "num_input_tokens_seen": 23456, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 52.75, |
| "learning_rate": 4.388888888888889e-05, |
| "loss": 0.1987, |
| "num_input_tokens_seen": 24992, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.9444444444444444, |
| "grad_norm": 1.2421875, |
| "learning_rate": 4.666666666666667e-05, |
| "loss": 0.177, |
| "num_input_tokens_seen": 26528, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 6.1875, |
| "learning_rate": 4.9444444444444446e-05, |
| "loss": 0.2215, |
| "num_input_tokens_seen": 28096, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.11327455192804337, |
| "eval_runtime": 0.8596, |
| "eval_samples_per_second": 46.533, |
| "eval_steps_per_second": 11.633, |
| "num_input_tokens_seen": 28096, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0555555555555556, |
| "grad_norm": 0.28515625, |
| "learning_rate": 4.9996991493233693e-05, |
| "loss": 0.0675, |
| "num_input_tokens_seen": 29632, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.93359375, |
| "learning_rate": 4.99847706754774e-05, |
| "loss": 0.1155, |
| "num_input_tokens_seen": 31200, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.1666666666666667, |
| "grad_norm": 6.21875, |
| "learning_rate": 4.9963154107272295e-05, |
| "loss": 0.0321, |
| "num_input_tokens_seen": 32768, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.2222222222222223, |
| "grad_norm": 3.015625, |
| "learning_rate": 4.993214991772563e-05, |
| "loss": 0.2483, |
| "num_input_tokens_seen": 34336, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.2777777777777777, |
| "grad_norm": 16.625, |
| "learning_rate": 4.989176976624511e-05, |
| "loss": 0.2041, |
| "num_input_tokens_seen": 35872, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 20.0, |
| "learning_rate": 4.9842028838154285e-05, |
| "loss": 0.0246, |
| "num_input_tokens_seen": 37440, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.3888888888888888, |
| "grad_norm": 2.71875, |
| "learning_rate": 4.978294583898196e-05, |
| "loss": 0.0045, |
| "num_input_tokens_seen": 38976, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.4444444444444444, |
| "grad_norm": 17.875, |
| "learning_rate": 4.971454298742779e-05, |
| "loss": 0.2086, |
| "num_input_tokens_seen": 40544, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 1.953125, |
| "learning_rate": 4.963684600700679e-05, |
| "loss": 0.1174, |
| "num_input_tokens_seen": 42144, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5, |
| "eval_loss": 0.12593017518520355, |
| "eval_runtime": 0.8603, |
| "eval_samples_per_second": 46.496, |
| "eval_steps_per_second": 11.624, |
| "num_input_tokens_seen": 42144, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 17.625, |
| "learning_rate": 4.9549884116375714e-05, |
| "loss": 0.2169, |
| "num_input_tokens_seen": 43744, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.6111111111111112, |
| "grad_norm": 3.296875, |
| "learning_rate": 4.9453690018345144e-05, |
| "loss": 0.1126, |
| "num_input_tokens_seen": 45280, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 0.173828125, |
| "learning_rate": 4.934829988758131e-05, |
| "loss": 0.1761, |
| "num_input_tokens_seen": 46880, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.7222222222222223, |
| "grad_norm": 59.25, |
| "learning_rate": 4.923375335700223e-05, |
| "loss": 0.4073, |
| "num_input_tokens_seen": 48480, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 3.1875, |
| "learning_rate": 4.9110093502873476e-05, |
| "loss": 0.0395, |
| "num_input_tokens_seen": 50080, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.8333333333333335, |
| "grad_norm": 25.625, |
| "learning_rate": 4.897736682860885e-05, |
| "loss": 0.1269, |
| "num_input_tokens_seen": 51648, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.8888888888888888, |
| "grad_norm": 13.1875, |
| "learning_rate": 4.883562324728241e-05, |
| "loss": 0.0354, |
| "num_input_tokens_seen": 53216, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.9444444444444444, |
| "grad_norm": 3.09375, |
| "learning_rate": 4.868491606285823e-05, |
| "loss": 0.1879, |
| "num_input_tokens_seen": 54752, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 6.21875, |
| "learning_rate": 4.8525301950144894e-05, |
| "loss": 0.1442, |
| "num_input_tokens_seen": 56352, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.11620314419269562, |
| "eval_runtime": 0.8614, |
| "eval_samples_per_second": 46.436, |
| "eval_steps_per_second": 11.609, |
| "num_input_tokens_seen": 56352, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0555555555555554, |
| "grad_norm": 0.0703125, |
| "learning_rate": 4.835684093348244e-05, |
| "loss": 0.0813, |
| "num_input_tokens_seen": 57888, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.111111111111111, |
| "grad_norm": 15.6875, |
| "learning_rate": 4.817959636416969e-05, |
| "loss": 0.1462, |
| "num_input_tokens_seen": 59488, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.1666666666666665, |
| "grad_norm": 0.08642578125, |
| "learning_rate": 4.7993634896640394e-05, |
| "loss": 0.1586, |
| "num_input_tokens_seen": 61024, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 38.25, |
| "learning_rate": 4.779902646339722e-05, |
| "loss": 0.0451, |
| "num_input_tokens_seen": 62528, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.2777777777777777, |
| "grad_norm": 89.5, |
| "learning_rate": 4.759584424871302e-05, |
| "loss": 0.1823, |
| "num_input_tokens_seen": 64064, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 0.8125, |
| "learning_rate": 4.7384164661109176e-05, |
| "loss": 0.2751, |
| "num_input_tokens_seen": 65696, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.388888888888889, |
| "grad_norm": 29.75, |
| "learning_rate": 4.7164067304621536e-05, |
| "loss": 0.135, |
| "num_input_tokens_seen": 67232, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 24.0, |
| "learning_rate": 4.693563494886455e-05, |
| "loss": 0.0955, |
| "num_input_tokens_seen": 68832, |
| "step": 220 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 60.75, |
| "learning_rate": 4.669895349790502e-05, |
| "loss": 0.1186, |
| "num_input_tokens_seen": 70432, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.11699075996875763, |
| "eval_runtime": 0.8611, |
| "eval_samples_per_second": 46.45, |
| "eval_steps_per_second": 11.613, |
| "num_input_tokens_seen": 70432, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5555555555555554, |
| "grad_norm": 5.46875, |
| "learning_rate": 4.645411195795709e-05, |
| "loss": 0.0087, |
| "num_input_tokens_seen": 72032, |
| "step": 230 |
| }, |
| { |
| "epoch": 2.611111111111111, |
| "grad_norm": 35.5, |
| "learning_rate": 4.620120240391065e-05, |
| "loss": 0.109, |
| "num_input_tokens_seen": 73632, |
| "step": 235 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 0.06884765625, |
| "learning_rate": 4.5940319944705736e-05, |
| "loss": 0.2209, |
| "num_input_tokens_seen": 75168, |
| "step": 240 |
| }, |
| { |
| "epoch": 2.7222222222222223, |
| "grad_norm": 19.25, |
| "learning_rate": 4.567156268756594e-05, |
| "loss": 0.0195, |
| "num_input_tokens_seen": 76768, |
| "step": 245 |
| }, |
| { |
| "epoch": 2.7777777777777777, |
| "grad_norm": 0.09033203125, |
| "learning_rate": 4.539503170110431e-05, |
| "loss": 0.0136, |
| "num_input_tokens_seen": 78336, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.8333333333333335, |
| "grad_norm": 0.1083984375, |
| "learning_rate": 4.5110830977315556e-05, |
| "loss": 0.0232, |
| "num_input_tokens_seen": 79872, |
| "step": 255 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 4.75, |
| "learning_rate": 4.4819067392468944e-05, |
| "loss": 0.0734, |
| "num_input_tokens_seen": 81440, |
| "step": 260 |
| }, |
| { |
| "epoch": 2.9444444444444446, |
| "grad_norm": 0.11376953125, |
| "learning_rate": 4.4519850666916484e-05, |
| "loss": 0.0539, |
| "num_input_tokens_seen": 83008, |
| "step": 265 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.66015625, |
| "learning_rate": 4.4213293323831585e-05, |
| "loss": 0.2272, |
| "num_input_tokens_seen": 84544, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.14363925158977509, |
| "eval_runtime": 0.8604, |
| "eval_samples_per_second": 46.493, |
| "eval_steps_per_second": 11.623, |
| "num_input_tokens_seen": 84544, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0555555555555554, |
| "grad_norm": 3.515625, |
| "learning_rate": 4.38995106468937e-05, |
| "loss": 0.0447, |
| "num_input_tokens_seen": 86112, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 0.2021484375, |
| "learning_rate": 4.357862063693486e-05, |
| "loss": 0.163, |
| "num_input_tokens_seen": 87680, |
| "step": 280 |
| }, |
| { |
| "epoch": 3.1666666666666665, |
| "grad_norm": 33.75, |
| "learning_rate": 4.325074396756437e-05, |
| "loss": 0.0236, |
| "num_input_tokens_seen": 89216, |
| "step": 285 |
| }, |
| { |
| "epoch": 3.2222222222222223, |
| "grad_norm": 0.212890625, |
| "learning_rate": 4.2916003939788403e-05, |
| "loss": 0.0395, |
| "num_input_tokens_seen": 90816, |
| "step": 290 |
| }, |
| { |
| "epoch": 3.2777777777777777, |
| "grad_norm": 0.40625, |
| "learning_rate": 4.257452643564155e-05, |
| "loss": 0.1351, |
| "num_input_tokens_seen": 92448, |
| "step": 295 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.8203125, |
| "learning_rate": 4.22264398708477e-05, |
| "loss": 0.0033, |
| "num_input_tokens_seen": 93984, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.388888888888889, |
| "grad_norm": 0.2109375, |
| "learning_rate": 4.1871875146528195e-05, |
| "loss": 0.1094, |
| "num_input_tokens_seen": 95584, |
| "step": 305 |
| }, |
| { |
| "epoch": 3.4444444444444446, |
| "grad_norm": 19.125, |
| "learning_rate": 4.1510965599975196e-05, |
| "loss": 0.0115, |
| "num_input_tokens_seen": 97088, |
| "step": 310 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 0.1572265625, |
| "learning_rate": 4.114384695450906e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 98688, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5, |
| "eval_loss": 0.15867777168750763, |
| "eval_runtime": 0.8711, |
| "eval_samples_per_second": 45.921, |
| "eval_steps_per_second": 11.48, |
| "num_input_tokens_seen": 98688, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 54.75, |
| "learning_rate": 4.077065726843828e-05, |
| "loss": 0.1914, |
| "num_input_tokens_seen": 100288, |
| "step": 320 |
| }, |
| { |
| "epoch": 3.611111111111111, |
| "grad_norm": 0.045166015625, |
| "learning_rate": 4.039153688314145e-05, |
| "loss": 0.0432, |
| "num_input_tokens_seen": 101824, |
| "step": 325 |
| }, |
| { |
| "epoch": 3.6666666666666665, |
| "grad_norm": 0.0947265625, |
| "learning_rate": 4.000662837029062e-05, |
| "loss": 0.0238, |
| "num_input_tokens_seen": 103392, |
| "step": 330 |
| }, |
| { |
| "epoch": 3.7222222222222223, |
| "grad_norm": 1.0859375, |
| "learning_rate": 3.961607647823583e-05, |
| "loss": 0.1576, |
| "num_input_tokens_seen": 104928, |
| "step": 335 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 30.125, |
| "learning_rate": 3.9220028077571295e-05, |
| "loss": 0.2329, |
| "num_input_tokens_seen": 106560, |
| "step": 340 |
| }, |
| { |
| "epoch": 3.8333333333333335, |
| "grad_norm": 0.0205078125, |
| "learning_rate": 3.881863210590332e-05, |
| "loss": 0.0114, |
| "num_input_tokens_seen": 108096, |
| "step": 345 |
| }, |
| { |
| "epoch": 3.888888888888889, |
| "grad_norm": 23.625, |
| "learning_rate": 3.841203951184095e-05, |
| "loss": 0.1721, |
| "num_input_tokens_seen": 109664, |
| "step": 350 |
| }, |
| { |
| "epoch": 3.9444444444444446, |
| "grad_norm": 0.0311279296875, |
| "learning_rate": 3.8000403198230387e-05, |
| "loss": 0.0934, |
| "num_input_tokens_seen": 111232, |
| "step": 355 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.031982421875, |
| "learning_rate": 3.75838779646545e-05, |
| "loss": 0.0823, |
| "num_input_tokens_seen": 112800, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.15528742969036102, |
| "eval_runtime": 0.8596, |
| "eval_samples_per_second": 46.535, |
| "eval_steps_per_second": 11.634, |
| "num_input_tokens_seen": 112800, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.055555555555555, |
| "grad_norm": 0.056884765625, |
| "learning_rate": 3.7162620449219e-05, |
| "loss": 0.1866, |
| "num_input_tokens_seen": 114400, |
| "step": 365 |
| }, |
| { |
| "epoch": 4.111111111111111, |
| "grad_norm": 82.5, |
| "learning_rate": 3.673678906964727e-05, |
| "loss": 0.1796, |
| "num_input_tokens_seen": 115936, |
| "step": 370 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 20.625, |
| "learning_rate": 3.630654396370594e-05, |
| "loss": 0.0363, |
| "num_input_tokens_seen": 117472, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 1.0234375, |
| "learning_rate": 3.5872046928983626e-05, |
| "loss": 0.0894, |
| "num_input_tokens_seen": 119040, |
| "step": 380 |
| }, |
| { |
| "epoch": 4.277777777777778, |
| "grad_norm": 13.5, |
| "learning_rate": 3.543346136204545e-05, |
| "loss": 0.0592, |
| "num_input_tokens_seen": 120640, |
| "step": 385 |
| }, |
| { |
| "epoch": 4.333333333333333, |
| "grad_norm": 0.48046875, |
| "learning_rate": 3.499095219698631e-05, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 122176, |
| "step": 390 |
| }, |
| { |
| "epoch": 4.388888888888889, |
| "grad_norm": 0.8828125, |
| "learning_rate": 3.454468584340588e-05, |
| "loss": 0.0325, |
| "num_input_tokens_seen": 123744, |
| "step": 395 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 0.287109375, |
| "learning_rate": 3.409483012382879e-05, |
| "loss": 0.0076, |
| "num_input_tokens_seen": 125280, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 0.1474609375, |
| "learning_rate": 3.364155421059342e-05, |
| "loss": 0.0133, |
| "num_input_tokens_seen": 126816, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.5, |
| "eval_loss": 0.15954655408859253, |
| "eval_runtime": 0.8629, |
| "eval_samples_per_second": 46.357, |
| "eval_steps_per_second": 11.589, |
| "num_input_tokens_seen": 126816, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.555555555555555, |
| "grad_norm": 10.125, |
| "learning_rate": 3.318502856223311e-05, |
| "loss": 0.0081, |
| "num_input_tokens_seen": 128352, |
| "step": 410 |
| }, |
| { |
| "epoch": 4.611111111111111, |
| "grad_norm": 0.059326171875, |
| "learning_rate": 3.272542485937369e-05, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 129920, |
| "step": 415 |
| }, |
| { |
| "epoch": 4.666666666666667, |
| "grad_norm": 0.0302734375, |
| "learning_rate": 3.2262915940171376e-05, |
| "loss": 0.0097, |
| "num_input_tokens_seen": 131424, |
| "step": 420 |
| }, |
| { |
| "epoch": 4.722222222222222, |
| "grad_norm": 0.0250244140625, |
| "learning_rate": 3.1797675735315455e-05, |
| "loss": 0.0343, |
| "num_input_tokens_seen": 132960, |
| "step": 425 |
| }, |
| { |
| "epoch": 4.777777777777778, |
| "grad_norm": 0.02880859375, |
| "learning_rate": 3.132987920262005e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 134528, |
| "step": 430 |
| }, |
| { |
| "epoch": 4.833333333333333, |
| "grad_norm": 0.031494140625, |
| "learning_rate": 3.085970226122962e-05, |
| "loss": 0.0016, |
| "num_input_tokens_seen": 136096, |
| "step": 435 |
| }, |
| { |
| "epoch": 4.888888888888889, |
| "grad_norm": 0.11962890625, |
| "learning_rate": 3.0387321725463e-05, |
| "loss": 0.0996, |
| "num_input_tokens_seen": 137728, |
| "step": 440 |
| }, |
| { |
| "epoch": 4.944444444444445, |
| "grad_norm": 42.5, |
| "learning_rate": 2.9912915238320754e-05, |
| "loss": 0.1518, |
| "num_input_tokens_seen": 139264, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 29.75, |
| "learning_rate": 2.9436661204680882e-05, |
| "loss": 0.0796, |
| "num_input_tokens_seen": 140800, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.16527244448661804, |
| "eval_runtime": 1.1028, |
| "eval_samples_per_second": 36.272, |
| "eval_steps_per_second": 9.068, |
| "num_input_tokens_seen": 140800, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.055555555555555, |
| "grad_norm": 0.1240234375, |
| "learning_rate": 2.8958738724208072e-05, |
| "loss": 0.0138, |
| "num_input_tokens_seen": 142400, |
| "step": 455 |
| }, |
| { |
| "epoch": 5.111111111111111, |
| "grad_norm": 0.267578125, |
| "learning_rate": 2.8479327524001636e-05, |
| "loss": 0.001, |
| "num_input_tokens_seen": 143872, |
| "step": 460 |
| }, |
| { |
| "epoch": 5.166666666666667, |
| "grad_norm": 0.0498046875, |
| "learning_rate": 2.7998607891007495e-05, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 145440, |
| "step": 465 |
| }, |
| { |
| "epoch": 5.222222222222222, |
| "grad_norm": 0.068359375, |
| "learning_rate": 2.7516760604219617e-05, |
| "loss": 0.0104, |
| "num_input_tokens_seen": 147040, |
| "step": 470 |
| }, |
| { |
| "epoch": 5.277777777777778, |
| "grad_norm": 0.181640625, |
| "learning_rate": 2.7033966866696457e-05, |
| "loss": 0.235, |
| "num_input_tokens_seen": 148576, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "grad_norm": 4.5625, |
| "learning_rate": 2.6550408237417885e-05, |
| "loss": 0.0356, |
| "num_input_tokens_seen": 150112, |
| "step": 480 |
| }, |
| { |
| "epoch": 5.388888888888889, |
| "grad_norm": 99.0, |
| "learning_rate": 2.6066266563008267e-05, |
| "loss": 0.1515, |
| "num_input_tokens_seen": 151712, |
| "step": 485 |
| }, |
| { |
| "epoch": 5.444444444444445, |
| "grad_norm": 7.25, |
| "learning_rate": 2.5581723909351406e-05, |
| "loss": 0.0724, |
| "num_input_tokens_seen": 153280, |
| "step": 490 |
| }, |
| { |
| "epoch": 5.5, |
| "grad_norm": 0.62890625, |
| "learning_rate": 2.5096962493123012e-05, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 154848, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.5, |
| "eval_loss": 0.17034828662872314, |
| "eval_runtime": 0.8647, |
| "eval_samples_per_second": 46.258, |
| "eval_steps_per_second": 11.565, |
| "num_input_tokens_seen": 154848, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.555555555555555, |
| "grad_norm": 0.0771484375, |
| "learning_rate": 2.461216461326642e-05, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 156352, |
| "step": 500 |
| }, |
| { |
| "epoch": 5.611111111111111, |
| "grad_norm": 43.25, |
| "learning_rate": 2.4127512582437485e-05, |
| "loss": 0.04, |
| "num_input_tokens_seen": 157920, |
| "step": 505 |
| }, |
| { |
| "epoch": 5.666666666666667, |
| "grad_norm": 14.25, |
| "learning_rate": 2.364318865844416e-05, |
| "loss": 0.0076, |
| "num_input_tokens_seen": 159488, |
| "step": 510 |
| }, |
| { |
| "epoch": 5.722222222222222, |
| "grad_norm": 4.125, |
| "learning_rate": 2.3159374975706884e-05, |
| "loss": 0.0055, |
| "num_input_tokens_seen": 161056, |
| "step": 515 |
| }, |
| { |
| "epoch": 5.777777777777778, |
| "grad_norm": 226.0, |
| "learning_rate": 2.2676253476765196e-05, |
| "loss": 0.0796, |
| "num_input_tokens_seen": 162624, |
| "step": 520 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 43.25, |
| "learning_rate": 2.2194005843856636e-05, |
| "loss": 0.0931, |
| "num_input_tokens_seen": 164128, |
| "step": 525 |
| }, |
| { |
| "epoch": 5.888888888888889, |
| "grad_norm": 141.0, |
| "learning_rate": 2.1712813430593436e-05, |
| "loss": 0.0651, |
| "num_input_tokens_seen": 165664, |
| "step": 530 |
| }, |
| { |
| "epoch": 5.944444444444445, |
| "grad_norm": 0.0517578125, |
| "learning_rate": 2.1232857193762924e-05, |
| "loss": 0.0026, |
| "num_input_tokens_seen": 167200, |
| "step": 535 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.6015625, |
| "learning_rate": 2.0754317625276983e-05, |
| "loss": 0.002, |
| "num_input_tokens_seen": 168736, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.17467355728149414, |
| "eval_runtime": 0.8602, |
| "eval_samples_per_second": 46.503, |
| "eval_steps_per_second": 11.626, |
| "num_input_tokens_seen": 168736, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.055555555555555, |
| "grad_norm": 0.07177734375, |
| "learning_rate": 2.02773746842965e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 170304, |
| "step": 545 |
| }, |
| { |
| "epoch": 6.111111111111111, |
| "grad_norm": 0.498046875, |
| "learning_rate": 1.980220772955602e-05, |
| "loss": 0.0132, |
| "num_input_tokens_seen": 171840, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.166666666666667, |
| "grad_norm": 0.2001953125, |
| "learning_rate": 1.932899545191433e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 173408, |
| "step": 555 |
| }, |
| { |
| "epoch": 6.222222222222222, |
| "grad_norm": 0.1259765625, |
| "learning_rate": 1.8857915807156092e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 174976, |
| "step": 560 |
| }, |
| { |
| "epoch": 6.277777777777778, |
| "grad_norm": 0.1669921875, |
| "learning_rate": 1.838914594906995e-05, |
| "loss": 0.0248, |
| "num_input_tokens_seen": 176544, |
| "step": 565 |
| }, |
| { |
| "epoch": 6.333333333333333, |
| "grad_norm": 0.10595703125, |
| "learning_rate": 1.792286216282824e-05, |
| "loss": 0.0479, |
| "num_input_tokens_seen": 178112, |
| "step": 570 |
| }, |
| { |
| "epoch": 6.388888888888889, |
| "grad_norm": 0.0654296875, |
| "learning_rate": 1.7459239798693364e-05, |
| "loss": 0.021, |
| "num_input_tokens_seen": 179584, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.444444444444445, |
| "grad_norm": 69.0, |
| "learning_rate": 1.699845320607571e-05, |
| "loss": 0.0222, |
| "num_input_tokens_seen": 181152, |
| "step": 580 |
| }, |
| { |
| "epoch": 6.5, |
| "grad_norm": 0.69140625, |
| "learning_rate": 1.6540675667967974e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 182688, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.5, |
| "eval_loss": 0.17897537350654602, |
| "eval_runtime": 0.8668, |
| "eval_samples_per_second": 46.148, |
| "eval_steps_per_second": 11.537, |
| "num_input_tokens_seen": 182688, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.555555555555555, |
| "grad_norm": 0.055419921875, |
| "learning_rate": 1.60860793357805e-05, |
| "loss": 0.0795, |
| "num_input_tokens_seen": 184224, |
| "step": 590 |
| }, |
| { |
| "epoch": 6.611111111111111, |
| "grad_norm": 1.625, |
| "learning_rate": 1.56348351646022e-05, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 185792, |
| "step": 595 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.0849609375, |
| "learning_rate": 1.5187112848911323e-05, |
| "loss": 0.1419, |
| "num_input_tokens_seen": 187392, |
| "step": 600 |
| }, |
| { |
| "epoch": 6.722222222222222, |
| "grad_norm": 0.3125, |
| "learning_rate": 1.47430807587603e-05, |
| "loss": 0.1698, |
| "num_input_tokens_seen": 188992, |
| "step": 605 |
| }, |
| { |
| "epoch": 6.777777777777778, |
| "grad_norm": 0.054931640625, |
| "learning_rate": 1.430290587645865e-05, |
| "loss": 0.0603, |
| "num_input_tokens_seen": 190560, |
| "step": 610 |
| }, |
| { |
| "epoch": 6.833333333333333, |
| "grad_norm": 0.4765625, |
| "learning_rate": 1.3866753733777765e-05, |
| "loss": 0.0022, |
| "num_input_tokens_seen": 192128, |
| "step": 615 |
| }, |
| { |
| "epoch": 6.888888888888889, |
| "grad_norm": 0.04736328125, |
| "learning_rate": 1.343478834970121e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 193664, |
| "step": 620 |
| }, |
| { |
| "epoch": 6.944444444444445, |
| "grad_norm": 0.84765625, |
| "learning_rate": 1.3007172168743854e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 195296, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 0.98828125, |
| "learning_rate": 1.2584065999863102e-05, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 196896, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.1758497655391693, |
| "eval_runtime": 0.8681, |
| "eval_samples_per_second": 46.079, |
| "eval_steps_per_second": 11.52, |
| "num_input_tokens_seen": 196896, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.055555555555555, |
| "grad_norm": 0.032958984375, |
| "learning_rate": 1.2165628955985314e-05, |
| "loss": 0.002, |
| "num_input_tokens_seen": 198528, |
| "step": 635 |
| }, |
| { |
| "epoch": 7.111111111111111, |
| "grad_norm": 0.345703125, |
| "learning_rate": 1.175201839416988e-05, |
| "loss": 0.0161, |
| "num_input_tokens_seen": 200096, |
| "step": 640 |
| }, |
| { |
| "epoch": 7.166666666666667, |
| "grad_norm": 0.07080078125, |
| "learning_rate": 1.1343389856433658e-05, |
| "loss": 0.2279, |
| "num_input_tokens_seen": 201696, |
| "step": 645 |
| }, |
| { |
| "epoch": 7.222222222222222, |
| "grad_norm": 3.484375, |
| "learning_rate": 1.0939897011258001e-05, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 203232, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.277777777777778, |
| "grad_norm": 144.0, |
| "learning_rate": 1.0541691595800337e-05, |
| "loss": 0.0375, |
| "num_input_tokens_seen": 204768, |
| "step": 655 |
| }, |
| { |
| "epoch": 7.333333333333333, |
| "grad_norm": 0.12109375, |
| "learning_rate": 1.0148923358832022e-05, |
| "loss": 0.0076, |
| "num_input_tokens_seen": 206368, |
| "step": 660 |
| }, |
| { |
| "epoch": 7.388888888888889, |
| "grad_norm": 0.890625, |
| "learning_rate": 9.761740004423927e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 207904, |
| "step": 665 |
| }, |
| { |
| "epoch": 7.444444444444445, |
| "grad_norm": 0.07666015625, |
| "learning_rate": 9.380287136401e-06, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 209472, |
| "step": 670 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 0.04248046875, |
| "learning_rate": 9.00470820358663e-06, |
| "loss": 0.0012, |
| "num_input_tokens_seen": 211072, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.5, |
| "eval_loss": 0.17746439576148987, |
| "eval_runtime": 0.8687, |
| "eval_samples_per_second": 46.046, |
| "eval_steps_per_second": 11.511, |
| "num_input_tokens_seen": 211072, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.555555555555555, |
| "grad_norm": 0.384765625, |
| "learning_rate": 8.635144445857406e-06, |
| "loss": 0.0217, |
| "num_input_tokens_seen": 212608, |
| "step": 680 |
| }, |
| { |
| "epoch": 7.611111111111111, |
| "grad_norm": 0.04931640625, |
| "learning_rate": 8.271734841028553e-06, |
| "loss": 0.0055, |
| "num_input_tokens_seen": 214176, |
| "step": 685 |
| }, |
| { |
| "epoch": 7.666666666666667, |
| "grad_norm": 0.119140625, |
| "learning_rate": 7.914616052590071e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 215744, |
| "step": 690 |
| }, |
| { |
| "epoch": 7.722222222222222, |
| "grad_norm": 74.5, |
| "learning_rate": 7.563922378313218e-06, |
| "loss": 0.0531, |
| "num_input_tokens_seen": 217248, |
| "step": 695 |
| }, |
| { |
| "epoch": 7.777777777777778, |
| "grad_norm": 0.06982421875, |
| "learning_rate": 7.219785699746573e-06, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 218816, |
| "step": 700 |
| }, |
| { |
| "epoch": 7.833333333333333, |
| "grad_norm": 0.0673828125, |
| "learning_rate": 6.882335432620779e-06, |
| "loss": 0.0862, |
| "num_input_tokens_seen": 220288, |
| "step": 705 |
| }, |
| { |
| "epoch": 7.888888888888889, |
| "grad_norm": 0.26171875, |
| "learning_rate": 6.55169847818059e-06, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 221856, |
| "step": 710 |
| }, |
| { |
| "epoch": 7.944444444444445, |
| "grad_norm": 0.390625, |
| "learning_rate": 6.22799917546252e-06, |
| "loss": 0.005, |
| "num_input_tokens_seen": 223424, |
| "step": 715 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 0.061767578125, |
| "learning_rate": 5.9113592545359945e-06, |
| "loss": 0.0293, |
| "num_input_tokens_seen": 225056, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.1754344254732132, |
| "eval_runtime": 0.8646, |
| "eval_samples_per_second": 46.262, |
| "eval_steps_per_second": 11.566, |
| "num_input_tokens_seen": 225056, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.055555555555555, |
| "grad_norm": 0.1865234375, |
| "learning_rate": 5.601897790725643e-06, |
| "loss": 0.1541, |
| "num_input_tokens_seen": 226656, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.11111111111111, |
| "grad_norm": 0.2490234375, |
| "learning_rate": 5.299731159831953e-06, |
| "loss": 0.0129, |
| "num_input_tokens_seen": 228160, |
| "step": 730 |
| }, |
| { |
| "epoch": 8.166666666666666, |
| "grad_norm": 0.1201171875, |
| "learning_rate": 5.004972994367102e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 229728, |
| "step": 735 |
| }, |
| { |
| "epoch": 8.222222222222221, |
| "grad_norm": 0.09375, |
| "learning_rate": 4.7177341408224e-06, |
| "loss": 0.0185, |
| "num_input_tokens_seen": 231328, |
| "step": 740 |
| }, |
| { |
| "epoch": 8.277777777777779, |
| "grad_norm": 0.0498046875, |
| "learning_rate": 4.438122617983443e-06, |
| "loss": 0.0768, |
| "num_input_tokens_seen": 232832, |
| "step": 745 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.0576171875, |
| "learning_rate": 4.166243576308712e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 234432, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.38888888888889, |
| "grad_norm": 0.267578125, |
| "learning_rate": 3.9021992583867325e-06, |
| "loss": 0.0514, |
| "num_input_tokens_seen": 235968, |
| "step": 755 |
| }, |
| { |
| "epoch": 8.444444444444445, |
| "grad_norm": 0.0751953125, |
| "learning_rate": 3.6460889604868626e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 237568, |
| "step": 760 |
| }, |
| { |
| "epoch": 8.5, |
| "grad_norm": 26.625, |
| "learning_rate": 3.398008995217988e-06, |
| "loss": 0.031, |
| "num_input_tokens_seen": 239168, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.5, |
| "eval_loss": 0.1779964417219162, |
| "eval_runtime": 0.863, |
| "eval_samples_per_second": 46.352, |
| "eval_steps_per_second": 11.588, |
| "num_input_tokens_seen": 239168, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.555555555555555, |
| "grad_norm": 0.22265625, |
| "learning_rate": 3.158052655309332e-06, |
| "loss": 0.0062, |
| "num_input_tokens_seen": 240800, |
| "step": 770 |
| }, |
| { |
| "epoch": 8.61111111111111, |
| "grad_norm": 0.07861328125, |
| "learning_rate": 2.9263101785268254e-06, |
| "loss": 0.0783, |
| "num_input_tokens_seen": 242368, |
| "step": 775 |
| }, |
| { |
| "epoch": 8.666666666666666, |
| "grad_norm": 0.04052734375, |
| "learning_rate": 2.7028687137384267e-06, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 243872, |
| "step": 780 |
| }, |
| { |
| "epoch": 8.722222222222221, |
| "grad_norm": 0.0859375, |
| "learning_rate": 2.487812288140945e-06, |
| "loss": 0.0055, |
| "num_input_tokens_seen": 245472, |
| "step": 785 |
| }, |
| { |
| "epoch": 8.777777777777779, |
| "grad_norm": 0.50390625, |
| "learning_rate": 2.281221775660894e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 247072, |
| "step": 790 |
| }, |
| { |
| "epoch": 8.833333333333334, |
| "grad_norm": 0.083984375, |
| "learning_rate": 2.0831748665410765e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 248640, |
| "step": 795 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "grad_norm": 0.03955078125, |
| "learning_rate": 1.893746038124497e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 250208, |
| "step": 800 |
| }, |
| { |
| "epoch": 8.944444444444445, |
| "grad_norm": 0.23046875, |
| "learning_rate": 1.713006526846439e-06, |
| "loss": 0.0121, |
| "num_input_tokens_seen": 251744, |
| "step": 805 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 2.203125, |
| "learning_rate": 1.541024301445404e-06, |
| "loss": 0.0176, |
| "num_input_tokens_seen": 253312, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.17836380004882812, |
| "eval_runtime": 0.8616, |
| "eval_samples_per_second": 46.423, |
| "eval_steps_per_second": 11.606, |
| "num_input_tokens_seen": 253312, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.055555555555555, |
| "grad_norm": 0.28125, |
| "learning_rate": 1.3778640374027985e-06, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 254944, |
| "step": 815 |
| }, |
| { |
| "epoch": 9.11111111111111, |
| "grad_norm": 0.04248046875, |
| "learning_rate": 1.2235870926211619e-06, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 256512, |
| "step": 820 |
| }, |
| { |
| "epoch": 9.166666666666666, |
| "grad_norm": 0.84765625, |
| "learning_rate": 1.0782514843499653e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 258016, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.222222222222221, |
| "grad_norm": 0.39453125, |
| "learning_rate": 9.419118673676924e-07, |
| "loss": 0.0673, |
| "num_input_tokens_seen": 259552, |
| "step": 830 |
| }, |
| { |
| "epoch": 9.277777777777779, |
| "grad_norm": 0.76953125, |
| "learning_rate": 8.146195134284052e-07, |
| "loss": 0.0256, |
| "num_input_tokens_seen": 261152, |
| "step": 835 |
| }, |
| { |
| "epoch": 9.333333333333334, |
| "grad_norm": 1.296875, |
| "learning_rate": 6.964222919805391e-07, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 262720, |
| "step": 840 |
| }, |
| { |
| "epoch": 9.38888888888889, |
| "grad_norm": 268.0, |
| "learning_rate": 5.87364652165176e-07, |
| "loss": 0.0971, |
| "num_input_tokens_seen": 264224, |
| "step": 845 |
| }, |
| { |
| "epoch": 9.444444444444445, |
| "grad_norm": 0.25390625, |
| "learning_rate": 4.874876061005173e-07, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 265792, |
| "step": 850 |
| }, |
| { |
| "epoch": 9.5, |
| "grad_norm": 1.109375, |
| "learning_rate": 3.9682871345891883e-07, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 267392, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.5, |
| "eval_loss": 0.17666800320148468, |
| "eval_runtime": 0.8648, |
| "eval_samples_per_second": 46.253, |
| "eval_steps_per_second": 11.563, |
| "num_input_tokens_seen": 267392, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.555555555555555, |
| "grad_norm": 4.0625, |
| "learning_rate": 3.1542206734221924e-07, |
| "loss": 0.0669, |
| "num_input_tokens_seen": 268896, |
| "step": 860 |
| }, |
| { |
| "epoch": 9.61111111111111, |
| "grad_norm": 0.431640625, |
| "learning_rate": 2.4329828146074095e-07, |
| "loss": 0.0059, |
| "num_input_tokens_seen": 270432, |
| "step": 865 |
| }, |
| { |
| "epoch": 9.666666666666666, |
| "grad_norm": 17.25, |
| "learning_rate": 1.8048447862070718e-07, |
| "loss": 0.0187, |
| "num_input_tokens_seen": 271968, |
| "step": 870 |
| }, |
| { |
| "epoch": 9.722222222222221, |
| "grad_norm": 0.061279296875, |
| "learning_rate": 1.2700428052447033e-07, |
| "loss": 0.0092, |
| "num_input_tokens_seen": 273504, |
| "step": 875 |
| }, |
| { |
| "epoch": 9.777777777777779, |
| "grad_norm": 0.04052734375, |
| "learning_rate": 8.28777988873486e-08, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 275104, |
| "step": 880 |
| }, |
| { |
| "epoch": 9.833333333333334, |
| "grad_norm": 24.375, |
| "learning_rate": 4.8121627874450625e-08, |
| "loss": 0.0067, |
| "num_input_tokens_seen": 276672, |
| "step": 885 |
| }, |
| { |
| "epoch": 9.88888888888889, |
| "grad_norm": 5.25, |
| "learning_rate": 2.2748837860270267e-08, |
| "loss": 0.002, |
| "num_input_tokens_seen": 278272, |
| "step": 890 |
| }, |
| { |
| "epoch": 9.944444444444445, |
| "grad_norm": 0.3359375, |
| "learning_rate": 6.768970513457151e-09, |
| "loss": 0.034, |
| "num_input_tokens_seen": 279808, |
| "step": 895 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.458984375, |
| "learning_rate": 1.8803520859811406e-10, |
| "loss": 0.1336, |
| "num_input_tokens_seen": 281408, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.1792178899049759, |
| "eval_runtime": 0.8626, |
| "eval_samples_per_second": 46.371, |
| "eval_steps_per_second": 11.593, |
| "num_input_tokens_seen": 281408, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 281408, |
| "step": 900, |
| "total_flos": 1.2671675052589056e+16, |
| "train_loss": 0.07668703520890428, |
| "train_runtime": 187.0339, |
| "train_samples_per_second": 19.248, |
| "train_steps_per_second": 4.812 |
| } |
| ], |
| "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": 1.2671675052589056e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
