{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.64,
  "eval_steps": 400,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "grad_norm": 3744.0,
      "learning_rate": 7.464101615137755e-08,
      "logits/chosen": -0.5232290029525757,
      "logits/rejected": -0.4501457214355469,
      "logps/chosen": -1.740447998046875,
      "logps/rejected": -1.7914221286773682,
      "loss": 105.9691,
      "rewards/accuracies": 0.45625001192092896,
      "rewards/chosen": -17.404481887817383,
      "rewards/margins": 0.5097407698631287,
      "rewards/rejected": -17.91421890258789,
      "step": 5
    },
    {
      "epoch": 0.32,
      "grad_norm": 1224.0,
      "learning_rate": 4.6945927106677224e-08,
      "logits/chosen": -0.4762292802333832,
      "logits/rejected": -0.37752610445022583,
      "logps/chosen": -1.6145172119140625,
      "logps/rejected": -1.8529552221298218,
      "loss": 87.966,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -16.145172119140625,
      "rewards/margins": 2.384378671646118,
      "rewards/rejected": -18.529550552368164,
      "step": 10
    },
    {
      "epoch": 0.48,
      "grad_norm": 1464.0,
      "learning_rate": 1.4288495612538426e-08,
      "logits/chosen": -0.5665463209152222,
      "logits/rejected": -0.5252197980880737,
      "logps/chosen": -1.716834306716919,
      "logps/rejected": -1.7479000091552734,
      "loss": 98.5938,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -17.16834259033203,
      "rewards/margins": 0.310658723115921,
      "rewards/rejected": -17.479000091552734,
      "step": 15
    },
    {
      "epoch": 0.64,
      "grad_norm": 4800.0,
      "learning_rate": 0.0,
      "logits/chosen": -0.5845416784286499,
      "logits/rejected": -0.5405458211898804,
      "logps/chosen": -1.8133652210235596,
      "logps/rejected": -1.8326694965362549,
      "loss": 101.0963,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -18.133655548095703,
      "rewards/margins": 0.19304139912128448,
      "rewards/rejected": -18.326696395874023,
      "step": 20
    },
    {
      "epoch": 0.64,
      "step": 20,
      "total_flos": 0.0,
      "train_loss": 98.40630645751953,
      "train_runtime": 107.1933,
      "train_samples_per_second": 5.971,
      "train_steps_per_second": 0.187
    }
  ],
  "logging_steps": 5,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000000,
  "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": 2,
  "trial_name": null,
  "trial_params": null
}