| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 100.0, |
| "global_step": 2226, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.22461814914645103, |
| "grad_norm": 10.72581958770752, |
| "learning_rate": 7.753818508535489e-07, |
| "logits/chosen": -2.3318123817443848, |
| "logits/rejected": -2.161390542984009, |
| "logps/chosen": -35.05125045776367, |
| "logps/rejected": -46.545249938964844, |
| "loss": 0.6777, |
| "rewards/accuracies": 0.5329999923706055, |
| "rewards/chosen": 0.002808334305882454, |
| "rewards/margins": 0.03920227661728859, |
| "rewards/rejected": -0.036407485604286194, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.44923629829290207, |
| "grad_norm": 9.643994331359863, |
| "learning_rate": 5.507637017070979e-07, |
| "logits/chosen": -2.329531192779541, |
| "logits/rejected": -2.1987500190734863, |
| "logps/chosen": -34.85200119018555, |
| "logps/rejected": -46.94874954223633, |
| "loss": 0.6625, |
| "rewards/accuracies": 0.6365000009536743, |
| "rewards/chosen": 0.005042862147092819, |
| "rewards/margins": 0.08301828056573868, |
| "rewards/rejected": -0.07795403897762299, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.6738544474393531, |
| "grad_norm": 8.894434928894043, |
| "learning_rate": 3.2614555256064693e-07, |
| "logits/chosen": -2.357734441757202, |
| "logits/rejected": -2.201906204223633, |
| "logps/chosen": -34.85974884033203, |
| "logps/rejected": -47.90024948120117, |
| "loss": 0.6458, |
| "rewards/accuracies": 0.6804999709129333, |
| "rewards/chosen": 0.003978912252932787, |
| "rewards/margins": 0.14209693670272827, |
| "rewards/rejected": -0.13809829950332642, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.8984725965858041, |
| "grad_norm": 11.70964241027832, |
| "learning_rate": 1.0152740341419585e-07, |
| "logits/chosen": -2.3715782165527344, |
| "logits/rejected": -2.221656322479248, |
| "logps/chosen": -35.111751556396484, |
| "logps/rejected": -47.67274856567383, |
| "loss": 0.6437, |
| "rewards/accuracies": 0.6812499761581421, |
| "rewards/chosen": -0.0028823164757341146, |
| "rewards/margins": 0.1492108404636383, |
| "rewards/rejected": -0.152044415473938, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2226, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "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": 0.0, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
