{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 4,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_logits/chosen": -3.926198720932007,
      "eval_logits/rejected": -3.8840649127960205,
      "eval_logps/chosen": -28.043203353881836,
      "eval_logps/rejected": -30.6968936920166,
      "eval_loss": 0.6933838129043579,
      "eval_rewards/accuracies": 0.42500001192092896,
      "eval_rewards/chosen": -0.0005649199592880905,
      "eval_rewards/margins": -0.00045604939805343747,
      "eval_rewards/rejected": -0.00010887047392316163,
      "eval_runtime": 2.307,
      "eval_samples_per_second": 86.692,
      "eval_steps_per_second": 86.692,
      "step": 2
    },
    {
      "epoch": 1.5,
      "eval_logits/chosen": -3.925924301147461,
      "eval_logits/rejected": -3.884033203125,
      "eval_logps/chosen": -28.0417423248291,
      "eval_logps/rejected": -30.701976776123047,
      "eval_loss": 0.6930555701255798,
      "eval_rewards/accuracies": 0.5149999856948853,
      "eval_rewards/chosen": -0.0004194255161564797,
      "eval_rewards/margins": 0.00019754555250983685,
      "eval_rewards/rejected": -0.0006169710541144013,
      "eval_runtime": 2.3043,
      "eval_samples_per_second": 86.795,
      "eval_steps_per_second": 86.795,
      "step": 3
    },
    {
      "epoch": 2.0,
      "eval_logits/chosen": -3.925567865371704,
      "eval_logits/rejected": -3.8829123973846436,
      "eval_logps/chosen": -28.049388885498047,
      "eval_logps/rejected": -30.710399627685547,
      "eval_loss": 0.6930195689201355,
      "eval_rewards/accuracies": 0.5149999856948853,
      "eval_rewards/chosen": -0.001183824148029089,
      "eval_rewards/margins": 0.00027557945577427745,
      "eval_rewards/rejected": -0.0014594033127650619,
      "eval_runtime": 2.2933,
      "eval_samples_per_second": 87.211,
      "eval_steps_per_second": 87.211,
      "step": 4
    }
  ],
  "logging_steps": 5,
  "max_steps": 4,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 64,
  "trial_name": null,
  "trial_params": null
}