| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 5.0, |
| "eval_steps": 500, |
| "global_step": 3220, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.7766990291262136, |
| "grad_norm": 0.9166448712348938, |
| "learning_rate": 4.99e-05, |
| "loss": 0.7466, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.20546381175518036, |
| "eval_runtime": 25.135, |
| "eval_samples_per_second": 19.893, |
| "eval_steps_per_second": 2.506, |
| "step": 644 |
| }, |
| { |
| "epoch": 1.553009708737864, |
| "grad_norm": 0.6259812116622925, |
| "learning_rate": 4.082720588235294e-05, |
| "loss": 0.2077, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.15930786728858948, |
| "eval_runtime": 25.1079, |
| "eval_samples_per_second": 19.914, |
| "eval_steps_per_second": 2.509, |
| "step": 1288 |
| }, |
| { |
| "epoch": 2.3293203883495144, |
| "grad_norm": 0.6516925692558289, |
| "learning_rate": 3.1636029411764704e-05, |
| "loss": 0.1524, |
| "step": 1500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.14231213927268982, |
| "eval_runtime": 25.1096, |
| "eval_samples_per_second": 19.913, |
| "eval_steps_per_second": 2.509, |
| "step": 1932 |
| }, |
| { |
| "epoch": 3.105631067961165, |
| "grad_norm": 0.6181855201721191, |
| "learning_rate": 2.2444852941176473e-05, |
| "loss": 0.1206, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.8823300970873786, |
| "grad_norm": 0.9624282717704773, |
| "learning_rate": 1.3253676470588237e-05, |
| "loss": 0.092, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.1376318633556366, |
| "eval_runtime": 25.1047, |
| "eval_samples_per_second": 19.917, |
| "eval_steps_per_second": 2.509, |
| "step": 2576 |
| }, |
| { |
| "epoch": 4.658640776699029, |
| "grad_norm": 0.9040796160697937, |
| "learning_rate": 4.0625000000000005e-06, |
| "loss": 0.083, |
| "step": 3000 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 3220, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.136133136384e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
