| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "global_step": 3100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 39.687320709228516, | |
| "eval_runtime": 2.8335, | |
| "eval_samples_per_second": 43.056, | |
| "eval_steps_per_second": 5.647, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 39.516841888427734, | |
| "eval_runtime": 2.9783, | |
| "eval_samples_per_second": 40.963, | |
| "eval_steps_per_second": 5.372, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 38.932273864746094, | |
| "eval_runtime": 3.1472, | |
| "eval_samples_per_second": 38.765, | |
| "eval_steps_per_second": 5.084, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 37.797115325927734, | |
| "eval_runtime": 3.1415, | |
| "eval_samples_per_second": 38.835, | |
| "eval_steps_per_second": 5.093, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 36.89667510986328, | |
| "eval_runtime": 3.1742, | |
| "eval_samples_per_second": 38.435, | |
| "eval_steps_per_second": 5.041, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 35.38439178466797, | |
| "eval_runtime": 3.1525, | |
| "eval_samples_per_second": 38.699, | |
| "eval_steps_per_second": 5.075, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 34.643463134765625, | |
| "eval_runtime": 3.1776, | |
| "eval_samples_per_second": 38.394, | |
| "eval_steps_per_second": 5.035, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 33.6444206237793, | |
| "eval_runtime": 3.1827, | |
| "eval_samples_per_second": 38.332, | |
| "eval_steps_per_second": 5.027, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 5.870967741935484e-08, | |
| "loss": 39.4735, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 32.891536712646484, | |
| "eval_runtime": 3.2424, | |
| "eval_samples_per_second": 37.627, | |
| "eval_steps_per_second": 4.935, | |
| "step": 558 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 31.138931274414062, | |
| "eval_runtime": 3.1172, | |
| "eval_samples_per_second": 39.138, | |
| "eval_steps_per_second": 5.133, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 30.020915985107422, | |
| "eval_runtime": 3.1861, | |
| "eval_samples_per_second": 38.291, | |
| "eval_steps_per_second": 5.022, | |
| "step": 682 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 28.798078536987305, | |
| "eval_runtime": 3.158, | |
| "eval_samples_per_second": 38.632, | |
| "eval_steps_per_second": 5.066, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 27.6728458404541, | |
| "eval_runtime": 3.1725, | |
| "eval_samples_per_second": 38.456, | |
| "eval_steps_per_second": 5.043, | |
| "step": 806 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 26.61111068725586, | |
| "eval_runtime": 3.1711, | |
| "eval_samples_per_second": 38.472, | |
| "eval_steps_per_second": 5.046, | |
| "step": 868 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 25.55615234375, | |
| "eval_runtime": 3.1645, | |
| "eval_samples_per_second": 38.553, | |
| "eval_steps_per_second": 5.056, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 24.428285598754883, | |
| "eval_runtime": 3.1894, | |
| "eval_samples_per_second": 38.251, | |
| "eval_steps_per_second": 5.017, | |
| "step": 992 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 4.741935483870968e-08, | |
| "loss": 29.1909, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 23.227224349975586, | |
| "eval_runtime": 3.3011, | |
| "eval_samples_per_second": 36.957, | |
| "eval_steps_per_second": 4.847, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 22.856584548950195, | |
| "eval_runtime": 3.1351, | |
| "eval_samples_per_second": 38.914, | |
| "eval_steps_per_second": 5.103, | |
| "step": 1116 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 22.174100875854492, | |
| "eval_runtime": 3.2113, | |
| "eval_samples_per_second": 37.99, | |
| "eval_steps_per_second": 4.982, | |
| "step": 1178 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 21.41921615600586, | |
| "eval_runtime": 3.1698, | |
| "eval_samples_per_second": 38.488, | |
| "eval_steps_per_second": 5.048, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 20.49114227294922, | |
| "eval_runtime": 3.202, | |
| "eval_samples_per_second": 38.101, | |
| "eval_steps_per_second": 4.997, | |
| "step": 1302 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 20.245670318603516, | |
| "eval_runtime": 3.1786, | |
| "eval_samples_per_second": 38.382, | |
| "eval_steps_per_second": 5.034, | |
| "step": 1364 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 19.58565902709961, | |
| "eval_runtime": 3.205, | |
| "eval_samples_per_second": 38.065, | |
| "eval_steps_per_second": 4.992, | |
| "step": 1426 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 19.287837982177734, | |
| "eval_runtime": 3.1902, | |
| "eval_samples_per_second": 38.242, | |
| "eval_steps_per_second": 5.015, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "learning_rate": 3.612903225806452e-08, | |
| "loss": 21.7917, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 18.806514739990234, | |
| "eval_runtime": 3.3778, | |
| "eval_samples_per_second": 36.118, | |
| "eval_steps_per_second": 4.737, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 18.336509704589844, | |
| "eval_runtime": 3.1436, | |
| "eval_samples_per_second": 38.81, | |
| "eval_steps_per_second": 5.09, | |
| "step": 1612 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 18.16900062561035, | |
| "eval_runtime": 3.2259, | |
| "eval_samples_per_second": 37.819, | |
| "eval_steps_per_second": 4.96, | |
| "step": 1674 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 17.55986213684082, | |
| "eval_runtime": 3.1926, | |
| "eval_samples_per_second": 38.213, | |
| "eval_steps_per_second": 5.012, | |
| "step": 1736 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 17.429519653320312, | |
| "eval_runtime": 3.19, | |
| "eval_samples_per_second": 38.245, | |
| "eval_steps_per_second": 5.016, | |
| "step": 1798 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 17.191831588745117, | |
| "eval_runtime": 3.2186, | |
| "eval_samples_per_second": 37.905, | |
| "eval_steps_per_second": 4.971, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 16.991268157958984, | |
| "eval_runtime": 3.2004, | |
| "eval_samples_per_second": 38.12, | |
| "eval_steps_per_second": 4.999, | |
| "step": 1922 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 16.737815856933594, | |
| "eval_runtime": 3.2002, | |
| "eval_samples_per_second": 38.122, | |
| "eval_steps_per_second": 5.0, | |
| "step": 1984 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 2.483870967741936e-08, | |
| "loss": 18.0017, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_loss": 16.585847854614258, | |
| "eval_runtime": 3.3866, | |
| "eval_samples_per_second": 36.024, | |
| "eval_steps_per_second": 4.725, | |
| "step": 2046 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 16.41578483581543, | |
| "eval_runtime": 3.1469, | |
| "eval_samples_per_second": 38.769, | |
| "eval_steps_per_second": 5.084, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 16.26023292541504, | |
| "eval_runtime": 3.2254, | |
| "eval_samples_per_second": 37.825, | |
| "eval_steps_per_second": 4.961, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 16.033435821533203, | |
| "eval_runtime": 3.1846, | |
| "eval_samples_per_second": 38.309, | |
| "eval_steps_per_second": 5.024, | |
| "step": 2232 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_loss": 16.080913543701172, | |
| "eval_runtime": 3.2091, | |
| "eval_samples_per_second": 38.017, | |
| "eval_steps_per_second": 4.986, | |
| "step": 2294 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 15.7175931930542, | |
| "eval_runtime": 3.1978, | |
| "eval_samples_per_second": 38.151, | |
| "eval_steps_per_second": 5.003, | |
| "step": 2356 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_loss": 15.824816703796387, | |
| "eval_runtime": 3.2153, | |
| "eval_samples_per_second": 37.944, | |
| "eval_steps_per_second": 4.976, | |
| "step": 2418 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 15.852823257446289, | |
| "eval_runtime": 3.1937, | |
| "eval_samples_per_second": 38.2, | |
| "eval_steps_per_second": 5.01, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 40.32, | |
| "learning_rate": 1.3548387096774193e-08, | |
| "loss": 16.124, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_loss": 15.77718734741211, | |
| "eval_runtime": 3.377, | |
| "eval_samples_per_second": 36.127, | |
| "eval_steps_per_second": 4.738, | |
| "step": 2542 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 15.797329902648926, | |
| "eval_runtime": 3.1524, | |
| "eval_samples_per_second": 38.701, | |
| "eval_steps_per_second": 5.075, | |
| "step": 2604 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 15.45718002319336, | |
| "eval_runtime": 3.2372, | |
| "eval_samples_per_second": 37.686, | |
| "eval_steps_per_second": 4.942, | |
| "step": 2666 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 15.622541427612305, | |
| "eval_runtime": 3.2026, | |
| "eval_samples_per_second": 38.094, | |
| "eval_steps_per_second": 4.996, | |
| "step": 2728 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_loss": 15.430057525634766, | |
| "eval_runtime": 3.1974, | |
| "eval_samples_per_second": 38.156, | |
| "eval_steps_per_second": 5.004, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 15.735359191894531, | |
| "eval_runtime": 3.2169, | |
| "eval_samples_per_second": 37.925, | |
| "eval_steps_per_second": 4.974, | |
| "step": 2852 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_loss": 15.529631614685059, | |
| "eval_runtime": 3.2177, | |
| "eval_samples_per_second": 37.916, | |
| "eval_steps_per_second": 4.973, | |
| "step": 2914 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 15.421342849731445, | |
| "eval_runtime": 3.225, | |
| "eval_samples_per_second": 37.829, | |
| "eval_steps_per_second": 4.961, | |
| "step": 2976 | |
| }, | |
| { | |
| "epoch": 48.39, | |
| "learning_rate": 2.2580645161290324e-09, | |
| "loss": 15.3787, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_loss": 15.512802124023438, | |
| "eval_runtime": 3.3809, | |
| "eval_samples_per_second": 36.085, | |
| "eval_steps_per_second": 4.732, | |
| "step": 3038 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 15.596056938171387, | |
| "eval_runtime": 3.151, | |
| "eval_samples_per_second": 38.718, | |
| "eval_steps_per_second": 5.078, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 3100, | |
| "total_flos": 1620261687859200.0, | |
| "train_loss": 23.064505024571574, | |
| "train_runtime": 2425.9558, | |
| "train_samples_per_second": 10.12, | |
| "train_steps_per_second": 1.278 | |
| } | |
| ], | |
| "max_steps": 3100, | |
| "num_train_epochs": 50, | |
| "total_flos": 1620261687859200.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |