| { | |
| "best_metric": 0.44799494293545944, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-9/checkpoint-3207", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 3207, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.453324739094023e-05, | |
| "loss": 0.5772, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.993745302655011e-05, | |
| "loss": 0.5332, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5163644552230835, | |
| "eval_matthews_correlation": 0.4043410378269883, | |
| "eval_runtime": 0.8124, | |
| "eval_samples_per_second": 1283.809, | |
| "eval_steps_per_second": 81.238, | |
| "step": 1069 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 3.5341658662159996e-05, | |
| "loss": 0.4009, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.0745864297769874e-05, | |
| "loss": 0.3795, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5874858498573303, | |
| "eval_matthews_correlation": 0.4092437141936731, | |
| "eval_runtime": 1.6757, | |
| "eval_samples_per_second": 622.409, | |
| "eval_steps_per_second": 39.385, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 2.6150069933379758e-05, | |
| "loss": 0.2679, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 2.1554275568989645e-05, | |
| "loss": 0.2669, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.8904589414596558, | |
| "eval_matthews_correlation": 0.44799494293545944, | |
| "eval_runtime": 0.7124, | |
| "eval_samples_per_second": 1464.018, | |
| "eval_steps_per_second": 92.642, | |
| "step": 3207 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5345, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 227521036404096.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 4.9129041755330344e-05, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 8, | |
| "seed": 38 | |
| } | |
| } | |