| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 12, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 8.833975791931152, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 76.2267, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.7217656373977661, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 41.8397, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 16.44710350036621, | |
| "learning_rate": 0.0001, | |
| "loss": 70.0907, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.9887499809265137, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 38.6177, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 15.259599685668945, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 65.9877, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.4123783111572266, | |
| "learning_rate": 0.0, | |
| "loss": 36.3016, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 12, | |
| "total_flos": 40668535028568.0, | |
| "train_loss": 54.844011306762695, | |
| "train_runtime": 47.4541, | |
| "train_samples_per_second": 4.215, | |
| "train_steps_per_second": 0.253 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 12, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 40668535028568.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |