| { | |
| "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_loss": 0.4282287061214447, | |
| "eval_runtime": 3.2319, | |
| "eval_samples_per_second": 959.178, | |
| "eval_steps_per_second": 20.112, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 0.861090898513794, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.6575, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.1691545695066452, | |
| "eval_runtime": 3.2816, | |
| "eval_samples_per_second": 944.667, | |
| "eval_steps_per_second": 19.808, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.09601032733917236, | |
| "eval_runtime": 3.3082, | |
| "eval_samples_per_second": 937.058, | |
| "eval_steps_per_second": 19.648, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "grad_norm": 0.8819959163665771, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.1958, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.07325955480337143, | |
| "eval_runtime": 3.4009, | |
| "eval_samples_per_second": 911.523, | |
| "eval_steps_per_second": 19.113, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 0.5379536151885986, | |
| "learning_rate": 1.1320754716981133e-06, | |
| "loss": 0.1126, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1590, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 391368939443328.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.31661862639116345, | |
| "num_train_epochs": 9, | |
| "temperature": 2 | |
| } | |
| } | |