| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.8094089264173703, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6031363088057901, | |
| "grad_norm": 1.831475853919983, | |
| "learning_rate": 1.3968636911942099e-05, | |
| "loss": 2.0688, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 16.3908, | |
| "eval_loss": 1.8676584959030151, | |
| "eval_rouge1": 39.017, | |
| "eval_rouge2": 16.1999, | |
| "eval_rougeL": 32.1156, | |
| "eval_rougeLsum": 35.7767, | |
| "eval_runtime": 165.5803, | |
| "eval_samples_per_second": 8.902, | |
| "eval_steps_per_second": 0.562, | |
| "step": 829 | |
| }, | |
| { | |
| "epoch": 1.2062726176115803, | |
| "grad_norm": 2.035099744796753, | |
| "learning_rate": 7.937273823884198e-06, | |
| "loss": 2.0394, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.8094089264173703, | |
| "grad_norm": 2.077877998352051, | |
| "learning_rate": 1.9059107358262969e-06, | |
| "loss": 2.0146, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1658, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 3137860544888832.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |