| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.008670019074041963, | |
| "eval_steps": 10, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0017340038148083927, | |
| "grad_norm": 11.838030815124512, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 12.8857, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0034680076296167853, | |
| "grad_norm": 0.8977920413017273, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 0.6798, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0034680076296167853, | |
| "eval_loss": 0.2337629497051239, | |
| "eval_runtime": 675.29, | |
| "eval_samples_per_second": 13.597, | |
| "eval_steps_per_second": 1.7, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0052020114444251775, | |
| "grad_norm": 0.34665364027023315, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.1571, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.006936015259233571, | |
| "grad_norm": 0.26853781938552856, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.1088, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.006936015259233571, | |
| "eval_loss": 0.09983003884553909, | |
| "eval_runtime": 686.1959, | |
| "eval_samples_per_second": 13.381, | |
| "eval_steps_per_second": 1.673, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008670019074041963, | |
| "grad_norm": 0.22648762166500092, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.0911, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1803863725327872.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |