| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.885496183206107, | |
| "eval_steps": 500, | |
| "global_step": 40, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12213740458015267, | |
| "grad_norm": 28.984829369986027, | |
| "learning_rate": 4.992293334332821e-06, | |
| "loss": 1.4068, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6106870229007634, | |
| "grad_norm": 5.240105914889918, | |
| "learning_rate": 4.809698831278217e-06, | |
| "loss": 0.6243, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.2213740458015268, | |
| "grad_norm": 1.2103549610602946, | |
| "learning_rate": 4.267766952966369e-06, | |
| "loss": 0.3896, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.83206106870229, | |
| "grad_norm": 1.2570422495768105, | |
| "learning_rate": 3.4567085809127247e-06, | |
| "loss": 0.2529, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.4427480916030535, | |
| "grad_norm": 1.1197951657629186, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2128, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.053435114503817, | |
| "grad_norm": 2.3662711159662524, | |
| "learning_rate": 1.5432914190872757e-06, | |
| "loss": 0.153, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.66412213740458, | |
| "grad_norm": 0.892855211805878, | |
| "learning_rate": 7.322330470336314e-07, | |
| "loss": 0.1002, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.2748091603053435, | |
| "grad_norm": 0.8045344023864228, | |
| "learning_rate": 1.9030116872178317e-07, | |
| "loss": 0.1044, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.885496183206107, | |
| "grad_norm": 0.7717421752636365, | |
| "learning_rate": 0.0, | |
| "loss": 0.0829, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.885496183206107, | |
| "step": 40, | |
| "total_flos": 15468790415360.0, | |
| "train_loss": 0.2595707342028618, | |
| "train_runtime": 1070.1995, | |
| "train_samples_per_second": 4.892, | |
| "train_steps_per_second": 0.037 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 40, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 15468790415360.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |