| { | |
| "best_metric": 0.4863630663948402, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-0/checkpoint-4276", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 4276, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 3.7270963191986084, | |
| "learning_rate": 1.0277702544991746e-05, | |
| "loss": 0.5849, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 16.02177619934082, | |
| "learning_rate": 9.408772671387439e-06, | |
| "loss": 0.5575, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 15.67810344696045, | |
| "learning_rate": 8.539842797783132e-06, | |
| "loss": 0.5369, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 19.78717041015625, | |
| "learning_rate": 7.670912924178823e-06, | |
| "loss": 0.5402, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5537068843841553, | |
| "eval_matthews_correlation": 0.472806497515492, | |
| "eval_runtime": 0.7082, | |
| "eval_samples_per_second": 1472.658, | |
| "eval_steps_per_second": 93.188, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 10.657214164733887, | |
| "learning_rate": 6.801983050574517e-06, | |
| "loss": 0.4772, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 9.260273933410645, | |
| "learning_rate": 5.933053176970209e-06, | |
| "loss": 0.4307, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 0.3461722433567047, | |
| "learning_rate": 5.064123303365902e-06, | |
| "loss": 0.4898, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 9.188729286193848, | |
| "learning_rate": 4.195193429761595e-06, | |
| "loss": 0.4323, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.8039517402648926, | |
| "eval_matthews_correlation": 0.4863630663948402, | |
| "eval_runtime": 0.7133, | |
| "eval_samples_per_second": 1462.131, | |
| "eval_steps_per_second": 92.522, | |
| "step": 4276 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6414, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 65242522366572.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.1146632418596053e-05, | |
| "num_train_epochs": 3, | |
| "per_device_train_batch_size": 4, | |
| "seed": 14 | |
| } | |
| } | |