| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 1670, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 6.208858013153076, | |
| "learning_rate": 3.973265321538069e-05, | |
| "loss": 3.8276, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 5.765829086303711, | |
| "learning_rate": 1.7364085661173347e-05, | |
| "loss": 3.4714, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 6.052618026733398, | |
| "learning_rate": 1.267564549716435e-06, | |
| "loss": 3.3029, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 1670, | |
| "total_flos": 850321843200000.0, | |
| "train_loss": 3.507169401431512, | |
| "train_runtime": 525.6016, | |
| "train_samples_per_second": 158.504, | |
| "train_steps_per_second": 3.177 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1670, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 850321843200000.0, | |
| "train_batch_size": 50, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |