| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0013227999265111152, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00014697776961234614, | |
| "grad_norm": 0.10356643795967102, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6371, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00014697776961234614, | |
| "eval_loss": NaN, | |
| "eval_runtime": 249.1633, | |
| "eval_samples_per_second": 11.498, | |
| "eval_steps_per_second": 5.751, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002939555392246923, | |
| "grad_norm": 0.10245563089847565, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6506, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0004409333088370384, | |
| "grad_norm": 0.13826023042201996, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9273, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0004409333088370384, | |
| "eval_loss": NaN, | |
| "eval_runtime": 247.1908, | |
| "eval_samples_per_second": 11.59, | |
| "eval_steps_per_second": 5.797, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005879110784493846, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.694, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0007348888480617307, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2828, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0008818666176740768, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.8857, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0008818666176740768, | |
| "eval_loss": NaN, | |
| "eval_runtime": 247.2998, | |
| "eval_samples_per_second": 11.585, | |
| "eval_steps_per_second": 5.795, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0010288443872864229, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 1.2798, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0011758221568987691, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.5146, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0013227999265111152, | |
| "grad_norm": 0.1688116192817688, | |
| "learning_rate": 0.00018, | |
| "loss": 0.8901, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0013227999265111152, | |
| "eval_loss": NaN, | |
| "eval_runtime": 247.1101, | |
| "eval_samples_per_second": 11.594, | |
| "eval_steps_per_second": 5.799, | |
| "step": 9 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5711923563724800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |