| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.046207497820401, | |
| "eval_steps": 500, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08718395815170009, | |
| "grad_norm": 1260.8033447265625, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 14.8341, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17436791630340018, | |
| "grad_norm": 745.888427734375, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 15.6171, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.26155187445510025, | |
| "grad_norm": 840.9533081054688, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 15.8498, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.34873583260680036, | |
| "grad_norm": 373.6918029785156, | |
| "learning_rate": 7.407407407407406e-07, | |
| "loss": 16.6859, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.43591979075850046, | |
| "grad_norm": 580.3277587890625, | |
| "learning_rate": 6.481481481481481e-07, | |
| "loss": 14.4898, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5231037489102005, | |
| "grad_norm": 77.99569702148438, | |
| "learning_rate": 5.555555555555555e-07, | |
| "loss": 15.7702, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6102877070619006, | |
| "grad_norm": 147.39584350585938, | |
| "learning_rate": 4.6296296296296297e-07, | |
| "loss": 15.0814, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.6974716652136007, | |
| "grad_norm": 141.38795471191406, | |
| "learning_rate": 3.703703703703703e-07, | |
| "loss": 12.6944, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7846556233653008, | |
| "grad_norm": 947.9025268554688, | |
| "learning_rate": 2.7777777777777776e-07, | |
| "loss": 12.4095, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.8718395815170009, | |
| "grad_norm": 862.9531860351562, | |
| "learning_rate": 1.8518518518518516e-07, | |
| "loss": 12.4597, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9590235396687009, | |
| "grad_norm": 247.12240600585938, | |
| "learning_rate": 9.259259259259258e-08, | |
| "loss": 11.3951, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 476.5635681152344, | |
| "eval_runtime": 22.3653, | |
| "eval_samples_per_second": 45.606, | |
| "eval_steps_per_second": 5.723, | |
| "step": 1147 | |
| }, | |
| { | |
| "epoch": 1.046207497820401, | |
| "grad_norm": 573.4383544921875, | |
| "learning_rate": 0.0, | |
| "loss": 12.2434, | |
| "step": 1200 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 600, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |