| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 26.25555680814376, | |
| "learning_rate": 5e-06, | |
| "loss": 1.483, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 17.39430141746882, | |
| "learning_rate": 9.330127018922195e-06, | |
| "loss": 1.1048, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 6.586149562982836, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.5822, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 4.531727627541418, | |
| "learning_rate": 1.7860619515673034e-06, | |
| "loss": 0.3652, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.7810551616818753, | |
| "learning_rate": 0.0, | |
| "loss": 0.2611, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 20, | |
| "total_flos": 523449139200.0, | |
| "train_loss": 0.5972473204135895, | |
| "train_runtime": 921.494, | |
| "train_samples_per_second": 0.152, | |
| "train_steps_per_second": 0.022 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 523449139200.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |