| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 7, | |
| "global_step": 132, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.030303030303030304, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.2128, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.21212121212121213, | |
| "grad_norm": 2.591930533784769, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 2.2112, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.21212121212121213, | |
| "eval_loss": 2.200307846069336, | |
| "eval_runtime": 142.2614, | |
| "eval_samples_per_second": 3.269, | |
| "eval_steps_per_second": 0.028, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.42424242424242425, | |
| "grad_norm": 3.209520918668776, | |
| "learning_rate": 2.1e-06, | |
| "loss": 2.1943, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.42424242424242425, | |
| "eval_loss": 2.147017478942871, | |
| "eval_runtime": 143.3053, | |
| "eval_samples_per_second": 3.245, | |
| "eval_steps_per_second": 0.028, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.6363636363636364, | |
| "grad_norm": 2.1132781771027593, | |
| "learning_rate": 2.5095609265912853e-06, | |
| "loss": 2.1573, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6363636363636364, | |
| "eval_loss": 2.0779454708099365, | |
| "eval_runtime": 143.4582, | |
| "eval_samples_per_second": 3.241, | |
| "eval_steps_per_second": 0.028, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8484848484848485, | |
| "grad_norm": 1.9056335733865675, | |
| "learning_rate": 1.3197749551783641e-06, | |
| "loss": 2.0595, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.8484848484848485, | |
| "eval_loss": 2.0294580459594727, | |
| "eval_runtime": 137.3982, | |
| "eval_samples_per_second": 3.384, | |
| "eval_steps_per_second": 0.029, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0606060606060606, | |
| "grad_norm": 2.1736242058284025, | |
| "learning_rate": 6.783887430182062e-07, | |
| "loss": 1.9961, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.0606060606060606, | |
| "eval_loss": 2.0057485103607178, | |
| "eval_runtime": 138.1845, | |
| "eval_samples_per_second": 3.365, | |
| "eval_steps_per_second": 0.029, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 3.1170870533174737, | |
| "learning_rate": 4.1931673730025623e-07, | |
| "loss": 1.9332, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "eval_loss": 1.9999465942382812, | |
| "eval_runtime": 139.0169, | |
| "eval_samples_per_second": 3.345, | |
| "eval_steps_per_second": 0.029, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.4848484848484849, | |
| "grad_norm": 2.023380630221034, | |
| "learning_rate": 2.1759855432049637e-07, | |
| "loss": 1.9101, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.4848484848484849, | |
| "eval_loss": 1.9939130544662476, | |
| "eval_runtime": 141.1776, | |
| "eval_samples_per_second": 3.294, | |
| "eval_steps_per_second": 0.028, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.696969696969697, | |
| "grad_norm": 2.6277704707226777, | |
| "learning_rate": 1.2154440189415328e-07, | |
| "loss": 1.906, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.696969696969697, | |
| "eval_loss": 1.990719199180603, | |
| "eval_runtime": 138.5039, | |
| "eval_samples_per_second": 3.357, | |
| "eval_steps_per_second": 0.029, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.9090909090909092, | |
| "grad_norm": 2.0677909551788045, | |
| "learning_rate": 7.843503292553053e-08, | |
| "loss": 1.9054, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.9090909090909092, | |
| "eval_loss": 1.9889289140701294, | |
| "eval_runtime": 142.8032, | |
| "eval_samples_per_second": 3.256, | |
| "eval_steps_per_second": 0.028, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.121212121212121, | |
| "grad_norm": 1.4792478673423035, | |
| "learning_rate": 6.038521136361391e-08, | |
| "loss": 1.9037, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.121212121212121, | |
| "eval_loss": 1.9878106117248535, | |
| "eval_runtime": 139.6825, | |
| "eval_samples_per_second": 3.329, | |
| "eval_steps_per_second": 0.029, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.3333333333333335, | |
| "grad_norm": 5.077301685002183, | |
| "learning_rate": 5.342647186003563e-08, | |
| "loss": 1.8786, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.3333333333333335, | |
| "eval_loss": 1.9871684312820435, | |
| "eval_runtime": 151.1266, | |
| "eval_samples_per_second": 3.077, | |
| "eval_steps_per_second": 0.026, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 1.6345215057101947, | |
| "learning_rate": 5.099824238664556e-08, | |
| "loss": 1.8962, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "eval_loss": 1.9865626096725464, | |
| "eval_runtime": 140.1891, | |
| "eval_samples_per_second": 3.317, | |
| "eval_steps_per_second": 0.029, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.757575757575758, | |
| "grad_norm": 1.7960257700653, | |
| "learning_rate": 5.024882880767712e-08, | |
| "loss": 1.8668, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.757575757575758, | |
| "eval_loss": 1.9858595132827759, | |
| "eval_runtime": 141.572, | |
| "eval_samples_per_second": 3.285, | |
| "eval_steps_per_second": 0.028, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.9696969696969697, | |
| "grad_norm": 2.888661327751688, | |
| "learning_rate": 5.0050722602692304e-08, | |
| "loss": 1.8988, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.9696969696969697, | |
| "eval_loss": 1.984999179840088, | |
| "eval_runtime": 139.6459, | |
| "eval_samples_per_second": 3.33, | |
| "eval_steps_per_second": 0.029, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.1818181818181817, | |
| "grad_norm": 1.5620758554693681, | |
| "learning_rate": 5.000789738737886e-08, | |
| "loss": 1.8966, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.1818181818181817, | |
| "eval_loss": 1.9841891527175903, | |
| "eval_runtime": 144.3125, | |
| "eval_samples_per_second": 3.222, | |
| "eval_steps_per_second": 0.028, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.393939393939394, | |
| "grad_norm": 2.0865509846490657, | |
| "learning_rate": 5.0000841090079794e-08, | |
| "loss": 1.8847, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.393939393939394, | |
| "eval_loss": 1.983519196510315, | |
| "eval_runtime": 143.6862, | |
| "eval_samples_per_second": 3.236, | |
| "eval_steps_per_second": 0.028, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.606060606060606, | |
| "grad_norm": 1.8436999654324047, | |
| "learning_rate": 5.000005037180778e-08, | |
| "loss": 1.8748, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 3.606060606060606, | |
| "eval_loss": 1.9828811883926392, | |
| "eval_runtime": 141.3381, | |
| "eval_samples_per_second": 3.29, | |
| "eval_steps_per_second": 0.028, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "grad_norm": 2.307936602188417, | |
| "learning_rate": 5.00000011344935e-08, | |
| "loss": 1.851, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "eval_loss": 1.9823122024536133, | |
| "eval_runtime": 142.0143, | |
| "eval_samples_per_second": 3.274, | |
| "eval_steps_per_second": 0.028, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 132, | |
| "total_flos": 130691559849984.0, | |
| "train_loss": 1.9532976656249075, | |
| "train_runtime": 25190.3368, | |
| "train_samples_per_second": 0.664, | |
| "train_steps_per_second": 0.005 | |
| } | |
| ], | |
| "logging_steps": 7, | |
| "max_steps": 132, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 27, | |
| "total_flos": 130691559849984.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |