| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.5536602700781805, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.9289267945984364e-05, | |
| "loss": 0.3176, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.857853589196873e-05, | |
| "loss": 0.146, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9562212252875999, | |
| "eval_f1": 0.7288810260946483, | |
| "eval_loss": 0.13318254053592682, | |
| "eval_precision": 0.6725223474543335, | |
| "eval_recall": 0.7955496299020735, | |
| "eval_runtime": 57.743, | |
| "eval_samples_per_second": 403.287, | |
| "eval_steps_per_second": 3.152, | |
| "step": 1407 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.7867803837953093e-05, | |
| "loss": 0.1295, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.715707178393746e-05, | |
| "loss": 0.1224, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 1.6446339729921822e-05, | |
| "loss": 0.1169, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9605602458582077, | |
| "eval_f1": 0.7573750499379718, | |
| "eval_loss": 0.11934245377779007, | |
| "eval_precision": 0.6978456292622442, | |
| "eval_recall": 0.8280079076824054, | |
| "eval_runtime": 57.4865, | |
| "eval_samples_per_second": 405.087, | |
| "eval_steps_per_second": 3.166, | |
| "step": 2814 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.5735607675906184e-05, | |
| "loss": 0.1119, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 1.5024875621890549e-05, | |
| "loss": 0.1095, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.4314143567874911e-05, | |
| "loss": 0.1069, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9633008348346227, | |
| "eval_f1": 0.7770200720535254, | |
| "eval_loss": 0.11019115895032883, | |
| "eval_precision": 0.7281459748402396, | |
| "eval_recall": 0.8329272217369317, | |
| "eval_runtime": 57.3128, | |
| "eval_samples_per_second": 406.314, | |
| "eval_steps_per_second": 3.176, | |
| "step": 4221 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.3603411513859277e-05, | |
| "loss": 0.1044, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 1.289267945984364e-05, | |
| "loss": 0.1038, | |
| "step": 5000 | |
| } | |
| ], | |
| "max_steps": 14070, | |
| "num_train_epochs": 10, | |
| "total_flos": 3387450405839616.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |