| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.84, | |
| "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": 1.070637822151184, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 9.4693, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.7139862775802612, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 9.1776, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.9775595664978027, | |
| "learning_rate": 0.0001, | |
| "loss": 8.7634, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 1.862234115600586, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 8.4361, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 2.3462889194488525, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 8.1063, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 1.7291364669799805, | |
| "learning_rate": 0.0, | |
| "loss": 7.9639, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "step": 12, | |
| "total_flos": 50821810811484.0, | |
| "train_loss": 8.652764320373535, | |
| "train_runtime": 311.9963, | |
| "train_samples_per_second": 0.628, | |
| "train_steps_per_second": 0.038 | |
| } | |
| ], | |
| "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": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 50821810811484.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |