| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.054910242872228086, | |
| "eval_steps": 10, | |
| "global_step": 26, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0021119324181626186, | |
| "eval_loss": 3.0706310272216797, | |
| "eval_runtime": 8.0649, | |
| "eval_samples_per_second": 24.799, | |
| "eval_steps_per_second": 12.399, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010559662090813094, | |
| "grad_norm": 1.2847050428390503, | |
| "learning_rate": 5e-05, | |
| "loss": 3.0647, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.021119324181626188, | |
| "grad_norm": 1.5589137077331543, | |
| "learning_rate": 0.0001, | |
| "loss": 2.6939, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021119324181626188, | |
| "eval_loss": 2.6773252487182617, | |
| "eval_runtime": 8.1789, | |
| "eval_samples_per_second": 24.453, | |
| "eval_steps_per_second": 12.227, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03167898627243928, | |
| "grad_norm": 2.162389039993286, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.4427, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.042238648363252376, | |
| "grad_norm": 1.5160914659500122, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.1306, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.042238648363252376, | |
| "eval_loss": 2.052976131439209, | |
| "eval_runtime": 8.2639, | |
| "eval_samples_per_second": 24.202, | |
| "eval_steps_per_second": 12.101, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05279831045406547, | |
| "grad_norm": 1.3696449995040894, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.9587, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2234632612085760.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |