| { | |
| "best_metric": 0.44397286177085654, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-0/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": 27.878910064697266, | |
| "learning_rate": 6.24294726518935e-06, | |
| "loss": 0.5804, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 19.8386287689209, | |
| "learning_rate": 4.337285706828742e-06, | |
| "loss": 0.5484, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 36.36659240722656, | |
| "learning_rate": 2.431624148468135e-06, | |
| "loss": 0.5157, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 30.039306640625, | |
| "learning_rate": 5.259625901075277e-07, | |
| "loss": 0.5107, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5229464769363403, | |
| "eval_matthews_correlation": 0.44397286177085654, | |
| "eval_runtime": 36.6034, | |
| "eval_samples_per_second": 28.495, | |
| "eval_steps_per_second": 1.803, | |
| "step": 2138 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2138, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 32664597685776.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 8.148608823549957e-06, | |
| "num_train_epochs": 1, | |
| "per_device_train_batch_size": 4, | |
| "seed": 10 | |
| } | |
| } | |