| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.24189646831156264, |
| "eval_steps": 500, |
| "global_step": 500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.9665795528888703, |
| "epoch": 0.04837929366231253, |
| "grad_norm": 4.359325885772705, |
| "learning_rate": 4.760522496371553e-05, |
| "loss": 1.7769, |
| "mean_token_accuracy": 0.5657356014847755, |
| "num_tokens": 50640.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.9170209395885467, |
| "epoch": 0.09675858732462506, |
| "grad_norm": 2.7793142795562744, |
| "learning_rate": 4.5186260280599906e-05, |
| "loss": 1.7405, |
| "mean_token_accuracy": 0.567837278842926, |
| "num_tokens": 99726.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.8398987126350403, |
| "epoch": 0.14513788098693758, |
| "grad_norm": 2.6684820652008057, |
| "learning_rate": 4.276729559748428e-05, |
| "loss": 1.6837, |
| "mean_token_accuracy": 0.5797487896680832, |
| "num_tokens": 150973.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.854783646464348, |
| "epoch": 0.1935171746492501, |
| "grad_norm": 2.2839877605438232, |
| "learning_rate": 4.0348330914368655e-05, |
| "loss": 1.6963, |
| "mean_token_accuracy": 0.5789705204963684, |
| "num_tokens": 203850.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.8420159757137298, |
| "epoch": 0.24189646831156264, |
| "grad_norm": 5.425283908843994, |
| "learning_rate": 3.792936623125303e-05, |
| "loss": 1.6699, |
| "mean_token_accuracy": 0.5854617989063263, |
| "num_tokens": 251974.0, |
| "step": 500 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 2067, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 215403548714496.0, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
