| { |
| "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": 31.791717529296875, |
| "learning_rate": 3.333333333333333e-07, |
| "loss": 1.1534664630889893, |
| "step": 1, |
| "token_acc": 0.7118055555555556 |
| }, |
| { |
| "epoch": 0.32388663967611336, |
| "grad_norm": 32.343624114990234, |
| "learning_rate": 9.95134034370785e-07, |
| "loss": 1.1402716636657715, |
| "step": 5, |
| "token_acc": 0.7121011615938833 |
| }, |
| { |
| "epoch": 0.6477732793522267, |
| "grad_norm": 29.438018798828125, |
| "learning_rate": 9.414737964294634e-07, |
| "loss": 0.9581957817077636, |
| "step": 10, |
| "token_acc": 0.7187350263536176 |
| }, |
| { |
| "epoch": 0.97165991902834, |
| "grad_norm": 10.981688499450684, |
| "learning_rate": 8.34565303179429e-07, |
| "loss": 0.5733880996704102, |
| "step": 15, |
| "token_acc": 0.801109537299339 |
| }, |
| { |
| "epoch": 1.2591093117408907, |
| "grad_norm": 4.334969520568848, |
| "learning_rate": 6.87303296707956e-07, |
| "loss": 0.5193193912506103, |
| "step": 20, |
| "token_acc": 0.8446018172100481 |
| }, |
| { |
| "epoch": 1.582995951417004, |
| "grad_norm": 4.081247806549072, |
| "learning_rate": 5.174497483512505e-07, |
| "loss": 0.4201364517211914, |
| "step": 25, |
| "token_acc": 0.8691533451746596 |
| }, |
| { |
| "epoch": 1.9068825910931175, |
| "grad_norm": 3.199093818664551, |
| "learning_rate": 3.454915028125263e-07, |
| "loss": 0.4021294116973877, |
| "step": 30, |
| "token_acc": 0.8769431588940311 |
| }, |
| { |
| "epoch": 2.194331983805668, |
| "grad_norm": 4.008501052856445, |
| "learning_rate": 1.9216926233717084e-07, |
| "loss": 0.4186545372009277, |
| "step": 35, |
| "token_acc": 0.866737176097303 |
| }, |
| { |
| "epoch": 2.5182186234817814, |
| "grad_norm": 3.1948792934417725, |
| "learning_rate": 7.597595192178702e-08, |
| "loss": 0.38708205223083497, |
| "step": 40, |
| "token_acc": 0.8758045292014303 |
| }, |
| { |
| "epoch": 2.8421052631578947, |
| "grad_norm": 2.8163161277770996, |
| "learning_rate": 1.0926199633097154e-08, |
| "loss": 0.37027928829193113, |
| "step": 45, |
| "token_acc": 0.8794045819281312 |
| } |
| ], |
| "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": 2.982229387824333e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
