| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 48, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.06477732793522267, |
| "grad_norm": 2.6590795516967773, |
| "learning_rate": 3.333333333333333e-07, |
| "loss": 0.7860867977142334, |
| "step": 1, |
| "token_acc": 0.7847862658856436 |
| }, |
| { |
| "epoch": 0.32388663967611336, |
| "grad_norm": 2.593881368637085, |
| "learning_rate": 9.95134034370785e-07, |
| "loss": 0.7774279117584229, |
| "step": 5, |
| "token_acc": 0.7862816580060082 |
| }, |
| { |
| "epoch": 0.6477732793522267, |
| "grad_norm": 2.2259292602539062, |
| "learning_rate": 9.414737964294634e-07, |
| "loss": 0.768182373046875, |
| "step": 10, |
| "token_acc": 0.787708951557998 |
| }, |
| { |
| "epoch": 0.97165991902834, |
| "grad_norm": 1.758042812347412, |
| "learning_rate": 8.34565303179429e-07, |
| "loss": 0.7460731506347656, |
| "step": 15, |
| "token_acc": 0.7901076130143491 |
| }, |
| { |
| "epoch": 1.2591093117408907, |
| "grad_norm": 1.3984026908874512, |
| "learning_rate": 6.87303296707956e-07, |
| "loss": 0.7199798583984375, |
| "step": 20, |
| "token_acc": 0.7959434966544631 |
| }, |
| { |
| "epoch": 1.582995951417004, |
| "grad_norm": 1.282883644104004, |
| "learning_rate": 5.174497483512505e-07, |
| "loss": 0.6806438446044922, |
| "step": 25, |
| "token_acc": 0.8018078572038763 |
| }, |
| { |
| "epoch": 1.9068825910931175, |
| "grad_norm": 1.15450119972229, |
| "learning_rate": 3.454915028125263e-07, |
| "loss": 0.6799572944641114, |
| "step": 30, |
| "token_acc": 0.8012063652516657 |
| }, |
| { |
| "epoch": 2.194331983805668, |
| "grad_norm": 0.977436363697052, |
| "learning_rate": 1.9216926233717084e-07, |
| "loss": 0.6614845752716064, |
| "step": 35, |
| "token_acc": 0.8086245801574344 |
| }, |
| { |
| "epoch": 2.5182186234817814, |
| "grad_norm": 0.9306056499481201, |
| "learning_rate": 7.597595192178702e-08, |
| "loss": 0.662669324874878, |
| "step": 40, |
| "token_acc": 0.8052767714763936 |
| }, |
| { |
| "epoch": 2.8421052631578947, |
| "grad_norm": 0.9451042413711548, |
| "learning_rate": 1.0926199633097154e-08, |
| "loss": 0.6642742156982422, |
| "step": 45, |
| "token_acc": 0.8049660845837912 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 48, |
| "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": 5.667590557631775e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
