| { | |
| "best_metric": 0.4098228742336231, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-4/checkpoint-2138", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 2138, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 5.885533332824707, | |
| "learning_rate": 6.456536809005447e-06, | |
| "loss": 0.6026, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 15.497381210327148, | |
| "learning_rate": 4.485676977196702e-06, | |
| "loss": 0.5242, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 21.243696212768555, | |
| "learning_rate": 2.514817145387958e-06, | |
| "loss": 0.5354, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 22.683589935302734, | |
| "learning_rate": 5.439573135792135e-07, | |
| "loss": 0.529, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5621954202651978, | |
| "eval_matthews_correlation": 0.4098228742336231, | |
| "eval_runtime": 0.739, | |
| "eval_samples_per_second": 1411.286, | |
| "eval_steps_per_second": 89.305, | |
| "step": 2138 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2138, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 32648039260944.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 8.427396640814191e-06, | |
| "num_train_epochs": 1, | |
| "per_device_train_batch_size": 4, | |
| "seed": 24 | |
| } | |
| } | |