| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 5.1328125, | |
| "learning_rate": 2.344043005375672e-05, | |
| "loss": 2.6022, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 3.90625, | |
| "learning_rate": 2.1877734716839608e-05, | |
| "loss": 2.3746, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 4.1328125, | |
| "learning_rate": 2.031503937992249e-05, | |
| "loss": 2.3571, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 4.625, | |
| "learning_rate": 1.8752344043005378e-05, | |
| "loss": 2.3713, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 3.01953125, | |
| "learning_rate": 1.7189648706088264e-05, | |
| "loss": 2.3099, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 3.98046875, | |
| "learning_rate": 1.5626953369171147e-05, | |
| "loss": 2.2603, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 16.125, | |
| "learning_rate": 1.4064258032254032e-05, | |
| "loss": 2.2645, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 3.755859375, | |
| "learning_rate": 1.2501562695336919e-05, | |
| "loss": 2.2809, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 3.244140625, | |
| "learning_rate": 1.0938867358419804e-05, | |
| "loss": 2.288, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 3.015625, | |
| "learning_rate": 9.376172021502689e-06, | |
| "loss": 2.262, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 5.66015625, | |
| "learning_rate": 7.813476684585574e-06, | |
| "loss": 2.2725, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 2.712890625, | |
| "learning_rate": 6.2507813476684595e-06, | |
| "loss": 2.2676, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 4.59765625, | |
| "learning_rate": 4.688086010751344e-06, | |
| "loss": 2.1962, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 7.265625, | |
| "learning_rate": 3.1253906738342297e-06, | |
| "loss": 2.2332, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 3.5390625, | |
| "learning_rate": 1.5626953369171149e-06, | |
| "loss": 2.2472, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.978515625, | |
| "learning_rate": 0.0, | |
| "loss": 2.2389, | |
| "step": 8000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 1.76841553870848e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |