| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 108, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.027777777777777776, | |
| "grad_norm": 0.2956169843673706, | |
| "learning_rate": 0.0, | |
| "loss": 2.8183, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 0.241030752658844, | |
| "learning_rate": 1.903846153846154e-05, | |
| "loss": 3.3956, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 0.34032657742500305, | |
| "learning_rate": 1.7115384615384617e-05, | |
| "loss": 3.591, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 0.4853314161300659, | |
| "learning_rate": 1.5192307692307693e-05, | |
| "loss": 3.576, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 0.46958625316619873, | |
| "learning_rate": 1.3269230769230769e-05, | |
| "loss": 3.4157, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3888888888888888, | |
| "grad_norm": 0.53203284740448, | |
| "learning_rate": 1.1346153846153847e-05, | |
| "loss": 3.3823, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.536034882068634, | |
| "learning_rate": 9.423076923076923e-06, | |
| "loss": 3.6569, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.9444444444444444, | |
| "grad_norm": 0.5476930737495422, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 3.2353, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.5972877144813538, | |
| "learning_rate": 5.576923076923077e-06, | |
| "loss": 3.3548, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.6853238940238953, | |
| "learning_rate": 3.653846153846154e-06, | |
| "loss": 3.4446, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 0.5809301733970642, | |
| "learning_rate": 1.7307692307692308e-06, | |
| "loss": 3.2611, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 108, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 4693397700870144.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |