| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 29.299820466786354, |
| "eval_steps": 500, |
| "global_step": 510, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 2.9299820466786355, |
| "grad_norm": 1.410841941833496, |
| "learning_rate": 4.917503719406088e-05, |
| "loss": 2.132, |
| "step": 51 |
| }, |
| { |
| "epoch": 5.859964093357271, |
| "grad_norm": 1.4861546754837036, |
| "learning_rate": 4.593820100105355e-05, |
| "loss": 1.5508, |
| "step": 102 |
| }, |
| { |
| "epoch": 8.789946140035907, |
| "grad_norm": 1.5175491571426392, |
| "learning_rate": 4.056969451220282e-05, |
| "loss": 1.0115, |
| "step": 153 |
| }, |
| { |
| "epoch": 11.719928186714542, |
| "grad_norm": 1.5818439722061157, |
| "learning_rate": 3.3616073079362926e-05, |
| "loss": 0.5528, |
| "step": 204 |
| }, |
| { |
| "epoch": 14.649910233393177, |
| "grad_norm": 1.2990632057189941, |
| "learning_rate": 2.578526897695321e-05, |
| "loss": 0.2567, |
| "step": 255 |
| }, |
| { |
| "epoch": 17.579892280071814, |
| "grad_norm": 0.6602067351341248, |
| "learning_rate": 1.7874518438250597e-05, |
| "loss": 0.1087, |
| "step": 306 |
| }, |
| { |
| "epoch": 20.50987432675045, |
| "grad_norm": 0.31337055563926697, |
| "learning_rate": 1.0689196860135234e-05, |
| "loss": 0.0477, |
| "step": 357 |
| }, |
| { |
| "epoch": 23.439856373429084, |
| "grad_norm": 0.3955017030239105, |
| "learning_rate": 4.960825378228082e-06, |
| "loss": 0.0299, |
| "step": 408 |
| }, |
| { |
| "epoch": 26.36983842010772, |
| "grad_norm": 0.2580316364765167, |
| "learning_rate": 1.2725963752426379e-06, |
| "loss": 0.0258, |
| "step": 459 |
| }, |
| { |
| "epoch": 29.299820466786354, |
| "grad_norm": 0.2776833474636078, |
| "learning_rate": 0.0, |
| "loss": 0.0242, |
| "step": 510 |
| } |
| ], |
| "logging_steps": 51, |
| "max_steps": 510, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 30, |
| "save_steps": 4, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.4841996401198694e+17, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
