| { | |
| "best_metric": 0.053496960550546646, | |
| "best_model_checkpoint": "/tmp/model/checkpoint-822", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 822, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.6867469879518073e-05, | |
| "loss": 1.7276, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.4939759036144585e-05, | |
| "loss": 0.4579, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.96617050067659e-05, | |
| "loss": 0.3088, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.76319350473613e-05, | |
| "loss": 0.35, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.56021650879567e-05, | |
| "loss": 0.2649, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.35723951285521e-05, | |
| "loss": 0.2502, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.15426251691475e-05, | |
| "loss": 0.3374, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 3.95128552097429e-05, | |
| "loss": 0.1944, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.7483085250338296e-05, | |
| "loss": 0.2182, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9739413680781759, | |
| "eval_auc": 0.9947087408402573, | |
| "eval_f1": 0.9719298245614035, | |
| "eval_loss": 0.09450644999742508, | |
| "eval_precision": 0.9736379613356766, | |
| "eval_recall": 0.9702276707530648, | |
| "eval_runtime": 22.7023, | |
| "eval_samples_per_second": 54.091, | |
| "eval_steps_per_second": 3.392, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 3.5453315290933695e-05, | |
| "loss": 0.2388, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.3423545331529095e-05, | |
| "loss": 0.1592, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.1393775372124494e-05, | |
| "loss": 0.1703, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 2.9364005412719893e-05, | |
| "loss": 0.2019, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 2.7334235453315295e-05, | |
| "loss": 0.2313, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.530446549391069e-05, | |
| "loss": 0.1362, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 2.327469553450609e-05, | |
| "loss": 0.1414, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 2.124492557510149e-05, | |
| "loss": 0.1077, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 1.9215155615696888e-05, | |
| "loss": 0.1434, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9723127035830619, | |
| "eval_auc": 0.9975742842139215, | |
| "eval_f1": 0.9707401032702238, | |
| "eval_loss": 0.08003830909729004, | |
| "eval_precision": 0.9543147208121827, | |
| "eval_recall": 0.9877408056042032, | |
| "eval_runtime": 23.305, | |
| "eval_samples_per_second": 52.692, | |
| "eval_steps_per_second": 3.304, | |
| "step": 548 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.7185385656292287e-05, | |
| "loss": 0.2318, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.5155615696887688e-05, | |
| "loss": 0.1714, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.3125845737483087e-05, | |
| "loss": 0.1123, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 1.1096075778078486e-05, | |
| "loss": 0.1017, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 9.066305818673885e-06, | |
| "loss": 0.1417, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 7.036535859269283e-06, | |
| "loss": 0.0908, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 5.006765899864682e-06, | |
| "loss": 0.1057, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.9769959404600813e-06, | |
| "loss": 0.1653, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 1.0148849797023005e-06, | |
| "loss": 0.0897, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9837133550488599, | |
| "eval_auc": 0.9982940020845161, | |
| "eval_f1": 0.9824561403508772, | |
| "eval_loss": 0.053496960550546646, | |
| "eval_precision": 0.984182776801406, | |
| "eval_recall": 0.9807355516637478, | |
| "eval_runtime": 22.6402, | |
| "eval_samples_per_second": 54.24, | |
| "eval_steps_per_second": 3.401, | |
| "step": 822 | |
| } | |
| ], | |
| "logging_steps": 30, | |
| "max_steps": 822, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 5.084249438613381e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |