| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.716981132075472, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6180645161290322, | |
| "eval_loss": 0.22989757359027863, | |
| "eval_runtime": 5.7081, | |
| "eval_samples_per_second": 543.089, | |
| "eval_steps_per_second": 11.387, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.3605, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8116129032258065, | |
| "eval_loss": 0.115962915122509, | |
| "eval_runtime": 5.9192, | |
| "eval_samples_per_second": 523.723, | |
| "eval_steps_per_second": 10.981, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8670967741935484, | |
| "eval_loss": 0.0799594596028328, | |
| "eval_runtime": 5.7417, | |
| "eval_samples_per_second": 539.909, | |
| "eval_steps_per_second": 11.321, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.1335, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8851612903225806, | |
| "eval_loss": 0.06618624925613403, | |
| "eval_runtime": 5.7358, | |
| "eval_samples_per_second": 540.469, | |
| "eval_steps_per_second": 11.332, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.1320754716981133e-06, | |
| "loss": 0.0936, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1590, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 390310534917408.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.3439719251489417, | |
| "num_train_epochs": 5, | |
| "temperature": 6 | |
| } | |
| } | |