| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9989344698987748, | |
| "eval_steps": 500, | |
| "global_step": 938, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21310602024507191, | |
| "grad_norm": 3.0930368900299072, | |
| "learning_rate": 7.092198581560284e-05, | |
| "loss": 2.0066, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.42621204049014383, | |
| "grad_norm": 1.5248420238494873, | |
| "learning_rate": 9.533965244865718e-05, | |
| "loss": 1.2202, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6393180607352158, | |
| "grad_norm": 1.2673641443252563, | |
| "learning_rate": 8.744075829383887e-05, | |
| "loss": 1.1874, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.8524240809802877, | |
| "grad_norm": 1.1734017133712769, | |
| "learning_rate": 7.954186413902054e-05, | |
| "loss": 1.1678, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0655301012253595, | |
| "grad_norm": 1.4848545789718628, | |
| "learning_rate": 7.164296998420222e-05, | |
| "loss": 1.0412, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2786361214704316, | |
| "grad_norm": 1.3286683559417725, | |
| "learning_rate": 6.374407582938389e-05, | |
| "loss": 0.8314, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.4917421417155035, | |
| "grad_norm": 1.1215488910675049, | |
| "learning_rate": 5.584518167456556e-05, | |
| "loss": 0.8177, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.7048481619605753, | |
| "grad_norm": 1.2876837253570557, | |
| "learning_rate": 4.794628751974723e-05, | |
| "loss": 0.8189, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.9179541822056474, | |
| "grad_norm": 1.2666414976119995, | |
| "learning_rate": 4.004739336492891e-05, | |
| "loss": 0.8247, | |
| "step": 900 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1407, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.902108727801119e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |