| { | |
| "best_metric": 0.8907678244972578, | |
| "best_model_checkpoint": "distilbert-hate_speech18\\run-4\\checkpoint-480", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 960, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 5.8838345155265075e-06, | |
| "loss": 0.549, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8820840950639853, | |
| "eval_loss": 0.4940944314002991, | |
| "eval_runtime": 8.0075, | |
| "eval_samples_per_second": 273.244, | |
| "eval_steps_per_second": 17.109, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.903195429605423e-06, | |
| "loss": 0.4705, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8907678244972578, | |
| "eval_loss": 0.4879714548587799, | |
| "eval_runtime": 7.9604, | |
| "eval_samples_per_second": 274.862, | |
| "eval_steps_per_second": 17.21, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.922556343684338e-06, | |
| "loss": 0.4471, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8893967093235832, | |
| "eval_loss": 0.4937082529067993, | |
| "eval_runtime": 8.0067, | |
| "eval_samples_per_second": 273.273, | |
| "eval_steps_per_second": 17.111, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 2.9419172577632537e-06, | |
| "loss": 0.4297, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8903107861060329, | |
| "eval_loss": 0.4932664930820465, | |
| "eval_runtime": 8.0508, | |
| "eval_samples_per_second": 271.775, | |
| "eval_steps_per_second": 17.017, | |
| "step": 960 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1680, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 651630374965248.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7641900951856909, | |
| "learning_rate": 6.864473601447592e-06, | |
| "num_train_epochs": 7, | |
| "per_device_eval_batch_size": 16, | |
| "per_device_train_batch_size": 32, | |
| "temperature": 23 | |
| } | |
| } | |