| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 2114, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23651844843897823, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 1.881740775780511e-05, | |
| "loss": 0.3019, | |
| "mean_token_accuracy": 0.9215967762470245, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.47303689687795647, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 1.763481551561022e-05, | |
| "loss": 0.246, | |
| "mean_token_accuracy": 0.9323436969518661, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7095553453169348, | |
| "grad_norm": 1.890625, | |
| "learning_rate": 1.6452223273415328e-05, | |
| "loss": 0.2338, | |
| "mean_token_accuracy": 0.9348658822774887, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9460737937559129, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 1.5269631031220437e-05, | |
| "loss": 0.2312, | |
| "mean_token_accuracy": 0.9352898219823838, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8456, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 1.4317446353092608e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |