| { | |
| "best_metric": 0.2045620264226157, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola\\run-0\\checkpoint-2138", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 2138, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 3.136430025100708, | |
| "learning_rate": 1.057670968337005e-06, | |
| "loss": 0.6141, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 3.4976818561553955, | |
| "learning_rate": 9.485201976107919e-07, | |
| "loss": 0.5835, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5739973187446594, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.1959, | |
| "eval_samples_per_second": 474.985, | |
| "eval_steps_per_second": 30.057, | |
| "step": 1069 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 5.231539249420166, | |
| "learning_rate": 8.393694268845788e-07, | |
| "loss": 0.5608, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 7.4941792488098145, | |
| "learning_rate": 7.302186561583657e-07, | |
| "loss": 0.5232, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5443820953369141, | |
| "eval_matthews_correlation": 0.2045620264226157, | |
| "eval_runtime": 0.9649, | |
| "eval_samples_per_second": 1080.926, | |
| "eval_steps_per_second": 68.4, | |
| "step": 2138 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5345, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 75421814032044.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.166821739063218e-06, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 8, | |
| "seed": 23 | |
| } | |
| } | |