| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.998331664998331, | |
| "global_step": 1870, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.8126709206927986e-06, | |
| "loss": 1.792, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.2750904902928595, | |
| "eval_f1_score": 0.17558198955755575, | |
| "eval_loss": 1.792249083518982, | |
| "eval_runtime": 351.3884, | |
| "eval_samples_per_second": 8.649, | |
| "eval_steps_per_second": 2.163, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.5569735642661806e-06, | |
| "loss": 1.7923, | |
| "step": 748 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.2750904902928595, | |
| "eval_f1_score": 0.15209869477487561, | |
| "eval_loss": 1.7973285913467407, | |
| "eval_runtime": 349.7493, | |
| "eval_samples_per_second": 8.689, | |
| "eval_steps_per_second": 2.173, | |
| "step": 748 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2.3012762078395626e-06, | |
| "loss": 1.7867, | |
| "step": 1122 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.39486673247778875, | |
| "eval_f1_score": 0.1978644849001168, | |
| "eval_loss": 1.773677110671997, | |
| "eval_runtime": 345.3478, | |
| "eval_samples_per_second": 8.8, | |
| "eval_steps_per_second": 2.201, | |
| "step": 1122 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 2.0455788514129442e-06, | |
| "loss": 1.7681, | |
| "step": 1496 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.49950641658440276, | |
| "eval_f1_score": 0.21909607145469126, | |
| "eval_loss": 1.751068115234375, | |
| "eval_runtime": 344.7764, | |
| "eval_samples_per_second": 8.814, | |
| "eval_steps_per_second": 2.204, | |
| "step": 1496 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.7898814949863265e-06, | |
| "loss": 1.7286, | |
| "step": 1870 | |
| } | |
| ], | |
| "max_steps": 4488, | |
| "num_train_epochs": 12, | |
| "total_flos": 9.102887348352e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |