| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 97.19626168224299, | |
| "eval_steps": 500, | |
| "global_step": 2600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 0.0003269230769230769, | |
| "loss": 1.6646, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 0.0006538461538461538, | |
| "loss": 1.517, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "learning_rate": 0.0009807692307692308, | |
| "loss": 1.4586, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 12.71, | |
| "learning_rate": 0.0009658119658119658, | |
| "loss": 1.3597, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 15.89, | |
| "learning_rate": 0.0009294871794871796, | |
| "loss": 1.2625, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 19.07, | |
| "learning_rate": 0.0008931623931623932, | |
| "loss": 1.1835, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 22.24, | |
| "learning_rate": 0.0008568376068376068, | |
| "loss": 1.1184, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 25.42, | |
| "learning_rate": 0.0008205128205128205, | |
| "loss": 1.0683, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 28.6, | |
| "learning_rate": 0.0007841880341880342, | |
| "loss": 1.0321, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 31.78, | |
| "learning_rate": 0.0007478632478632479, | |
| "loss": 1.0009, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 34.95, | |
| "learning_rate": 0.0007115384615384616, | |
| "loss": 0.9663, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 38.13, | |
| "learning_rate": 0.0006752136752136753, | |
| "loss": 0.938, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 41.31, | |
| "learning_rate": 0.0006388888888888888, | |
| "loss": 0.9178, | |
| "step": 1105 | |
| }, | |
| { | |
| "epoch": 44.49, | |
| "learning_rate": 0.0006025641025641026, | |
| "loss": 0.8984, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 47.66, | |
| "learning_rate": 0.0005662393162393163, | |
| "loss": 0.8827, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 50.84, | |
| "learning_rate": 0.0005299145299145299, | |
| "loss": 0.8693, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 54.02, | |
| "learning_rate": 0.0004935897435897436, | |
| "loss": 0.8514, | |
| "step": 1445 | |
| }, | |
| { | |
| "epoch": 57.2, | |
| "learning_rate": 0.0004572649572649573, | |
| "loss": 0.8349, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 60.37, | |
| "learning_rate": 0.00042094017094017095, | |
| "loss": 0.824, | |
| "step": 1615 | |
| }, | |
| { | |
| "epoch": 63.55, | |
| "learning_rate": 0.00038461538461538467, | |
| "loss": 0.8138, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 66.73, | |
| "learning_rate": 0.0003482905982905983, | |
| "loss": 0.8044, | |
| "step": 1785 | |
| }, | |
| { | |
| "epoch": 69.91, | |
| "learning_rate": 0.00031196581196581195, | |
| "loss": 0.7962, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 73.08, | |
| "learning_rate": 0.0002756410256410257, | |
| "loss": 0.7854, | |
| "step": 1955 | |
| }, | |
| { | |
| "epoch": 76.26, | |
| "learning_rate": 0.00023931623931623932, | |
| "loss": 0.7756, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 79.44, | |
| "learning_rate": 0.000202991452991453, | |
| "loss": 0.7663, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 82.62, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 0.7569, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 85.79, | |
| "learning_rate": 0.00013034188034188036, | |
| "loss": 0.7491, | |
| "step": 2295 | |
| }, | |
| { | |
| "epoch": 88.97, | |
| "learning_rate": 9.401709401709401e-05, | |
| "loss": 0.7427, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 92.15, | |
| "learning_rate": 5.76923076923077e-05, | |
| "loss": 0.7344, | |
| "step": 2465 | |
| }, | |
| { | |
| "epoch": 95.33, | |
| "learning_rate": 2.1367521367521368e-05, | |
| "loss": 0.7307, | |
| "step": 2550 | |
| } | |
| ], | |
| "logging_steps": 85, | |
| "max_steps": 2600, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 2.7054682298346635e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |