{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0021746221593998043,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005436555398499511,
      "grad_norm": 13.953805923461914,
      "learning_rate": 1.4563106796116505e-07,
      "logits/chosen": 0.7499047517776489,
      "logits/rejected": 0.7857828140258789,
      "logps/chosen": -277.1124267578125,
      "logps/rejected": -273.15863037109375,
      "loss": 0.6945,
      "rewards/accuracies": 0.2874999940395355,
      "rewards/chosen": 0.0018529916414991021,
      "rewards/margins": -0.0021975873969495296,
      "rewards/rejected": 0.00405057892203331,
      "step": 5
    },
    {
      "epoch": 0.0010873110796999022,
      "grad_norm": 12.302263259887695,
      "learning_rate": 3.276699029126214e-07,
      "logits/chosen": 0.6590738296508789,
      "logits/rejected": 0.6296309232711792,
      "logps/chosen": -274.404296875,
      "logps/rejected": -226.207763671875,
      "loss": 0.6908,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.0032993126660585403,
      "rewards/margins": 0.005577206611633301,
      "rewards/rejected": -0.0022778944112360477,
      "step": 10
    },
    {
      "epoch": 0.0016309666195498532,
      "grad_norm": 12.561519622802734,
      "learning_rate": 5.097087378640777e-07,
      "logits/chosen": 0.7245112061500549,
      "logits/rejected": 0.6608446836471558,
      "logps/chosen": -273.90106201171875,
      "logps/rejected": -254.3201141357422,
      "loss": 0.693,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": 0.0016085291281342506,
      "rewards/margins": 0.000873942393809557,
      "rewards/rejected": 0.0007345870253629982,
      "step": 15
    },
    {
      "epoch": 0.0021746221593998043,
      "grad_norm": 10.14516830444336,
      "learning_rate": 6.91747572815534e-07,
      "logits/chosen": 0.7335291504859924,
      "logits/rejected": 0.6993385553359985,
      "logps/chosen": -273.7506103515625,
      "logps/rejected": -292.1577453613281,
      "loss": 0.6889,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.0034298517275601625,
      "rewards/margins": 0.0093202730640769,
      "rewards/rejected": -0.005890421569347382,
      "step": 20
    },
    {
      "epoch": 0.0021746221593998043,
      "eval_logits/chosen": 0.6152238249778748,
      "eval_logits/rejected": 0.4639601409435272,
      "eval_logps/chosen": -275.6947937011719,
      "eval_logps/rejected": -254.22560119628906,
      "eval_loss": 0.6924488544464111,
      "eval_rewards/accuracies": 0.4749999940395355,
      "eval_rewards/chosen": 0.006965952459722757,
      "eval_rewards/margins": 0.0024989857338368893,
      "eval_rewards/rejected": 0.004466966260224581,
      "eval_runtime": 35.1741,
      "eval_samples_per_second": 5.686,
      "eval_steps_per_second": 5.686,
      "step": 20
    }
  ],
  "logging_steps": 5,
  "max_steps": 1032,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 206.0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}