| { |
| "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": 2.664837121963501, |
| "learning_rate": 3.333333333333333e-07, |
| "loss": 0.7869457006454468, |
| "step": 1, |
| "token_acc": 0.7843859578004337 |
| }, |
| { |
| "epoch": 0.32388663967611336, |
| "grad_norm": 2.596522092819214, |
| "learning_rate": 9.95134034370785e-07, |
| "loss": 0.7780666351318359, |
| "step": 5, |
| "token_acc": 0.7861042545117468 |
| }, |
| { |
| "epoch": 0.6477732793522267, |
| "grad_norm": 2.233597755432129, |
| "learning_rate": 9.414737964294634e-07, |
| "loss": 0.7686914443969727, |
| "step": 10, |
| "token_acc": 0.7876514979589609 |
| }, |
| { |
| "epoch": 0.97165991902834, |
| "grad_norm": 1.7652353048324585, |
| "learning_rate": 8.34565303179429e-07, |
| "loss": 0.7461651802062989, |
| "step": 15, |
| "token_acc": 0.79009821121334 |
| }, |
| { |
| "epoch": 1.2591093117408907, |
| "grad_norm": 1.3956773281097412, |
| "learning_rate": 6.87303296707956e-07, |
| "loss": 0.7195645809173584, |
| "step": 20, |
| "token_acc": 0.7960847975148048 |
| }, |
| { |
| "epoch": 1.582995951417004, |
| "grad_norm": 1.2905428409576416, |
| "learning_rate": 5.174497483512505e-07, |
| "loss": 0.680579948425293, |
| "step": 25, |
| "token_acc": 0.8018313261155477 |
| }, |
| { |
| "epoch": 1.9068825910931175, |
| "grad_norm": 1.1578842401504517, |
| "learning_rate": 3.454915028125263e-07, |
| "loss": 0.6798376083374024, |
| "step": 30, |
| "token_acc": 0.8013003982690089 |
| }, |
| { |
| "epoch": 2.194331983805668, |
| "grad_norm": 0.9808235764503479, |
| "learning_rate": 1.9216926233717084e-07, |
| "loss": 0.6614072322845459, |
| "step": 35, |
| "token_acc": 0.8087056890617363 |
| }, |
| { |
| "epoch": 2.5182186234817814, |
| "grad_norm": 0.9294739961624146, |
| "learning_rate": 7.597595192178702e-08, |
| "loss": 0.6624435424804688, |
| "step": 40, |
| "token_acc": 0.8052142003362278 |
| }, |
| { |
| "epoch": 2.8421052631578947, |
| "grad_norm": 0.9468541741371155, |
| "learning_rate": 1.0926199633097154e-08, |
| "loss": 0.6641257286071778, |
| "step": 45, |
| "token_acc": 0.8050948131509915 |
| } |
| ], |
| "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 |
| } |
|
|
