| { | |
| "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": 8.356196403503418, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 76.257, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.7907499074935913, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 41.8413, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 16.30155372619629, | |
| "learning_rate": 0.0001, | |
| "loss": 70.232, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.569784164428711, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 38.5374, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 15.395219802856445, | |
| "learning_rate": 5e-05, | |
| "loss": 66.8565, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.543508529663086, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 36.567, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 12, | |
| "total_flos": 40557671234760.0, | |
| "train_loss": 55.04854393005371, | |
| "train_runtime": 49.2052, | |
| "train_samples_per_second": 4.065, | |
| "train_steps_per_second": 0.244 | |
| } | |
| ], | |
| "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": 40557671234760.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |