| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.7906976744186047, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 9363.44167827181, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 30.9126, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 1160.9061028219837, | |
| "learning_rate": 3.946089741159648e-05, | |
| "loss": 24.2863, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 1517.0566103485007, | |
| "learning_rate": 3.372483275737468e-05, | |
| "loss": 34.3937, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 795.2235520034859, | |
| "learning_rate": 2.3472963553338614e-05, | |
| "loss": 24.2892, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 458.09743259288496, | |
| "learning_rate": 1.2078404679216864e-05, | |
| "loss": 19.4514, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 299.7088111243142, | |
| "learning_rate": 3.290243771741275e-06, | |
| "loss": 17.1994, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 275.3311095749166, | |
| "learning_rate": 0.0, | |
| "loss": 15.8193, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "step": 30, | |
| "total_flos": 8193589641216.0, | |
| "train_loss": 22.794086201985678, | |
| "train_runtime": 180.5723, | |
| "train_samples_per_second": 22.661, | |
| "train_steps_per_second": 0.166 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 8193589641216.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |