| { | |
| "best_metric": 0.9878331402085747, | |
| "best_model_checkpoint": "./outputs/mdeberta-v3-base-output/cometa_8_0.01_0.00005_09-08-22_17-20/checkpoint-1456", | |
| "epoch": 4.0, | |
| "global_step": 1456, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.991228070175439e-05, | |
| "loss": 0.0899, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9917505892436255, | |
| "eval_f1": 0.744355909694555, | |
| "eval_loss": 0.022882504388689995, | |
| "eval_precision": 0.8629715165511932, | |
| "eval_recall": 0.6544074722708698, | |
| "eval_runtime": 12.5601, | |
| "eval_samples_per_second": 231.447, | |
| "eval_steps_per_second": 28.981, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.6608187134502928e-05, | |
| "loss": 0.0262, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9977715877437325, | |
| "eval_f1": 0.9390029325513196, | |
| "eval_loss": 0.008390289731323719, | |
| "eval_precision": 0.943429581614614, | |
| "eval_recall": 0.9346176298890835, | |
| "eval_runtime": 12.5029, | |
| "eval_samples_per_second": 232.507, | |
| "eval_steps_per_second": 29.113, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.3304093567251464e-05, | |
| "loss": 0.0102, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9990464966788086, | |
| "eval_f1": 0.9741654571843252, | |
| "eval_loss": 0.0033252816647291183, | |
| "eval_precision": 0.9688221709006929, | |
| "eval_recall": 0.9795680093403386, | |
| "eval_runtime": 12.4462, | |
| "eval_samples_per_second": 233.565, | |
| "eval_steps_per_second": 29.246, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0046, | |
| "step": 1456 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9995500321405614, | |
| "eval_f1": 0.9878331402085747, | |
| "eval_loss": 0.0015423308359459043, | |
| "eval_precision": 0.980448533640023, | |
| "eval_recall": 0.9953298307063632, | |
| "eval_runtime": 12.5941, | |
| "eval_samples_per_second": 230.823, | |
| "eval_steps_per_second": 28.902, | |
| "step": 1456 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1456, | |
| "total_flos": 613160172693882.0, | |
| "train_loss": 0.032722131645941475, | |
| "train_runtime": 391.1103, | |
| "train_samples_per_second": 29.731, | |
| "train_steps_per_second": 3.723 | |
| } | |
| ], | |
| "max_steps": 1456, | |
| "num_train_epochs": 4, | |
| "total_flos": 613160172693882.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |