{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "eval_logits/chosen": -2.8025150299072266,
      "eval_logits/rejected": -3.1548430919647217,
      "eval_logps/chosen": -285.6490478515625,
      "eval_logps/rejected": -73.34615325927734,
      "eval_loss": 1.696915387583431e-05,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 11.185397148132324,
      "eval_rewards/margins": 12.019237518310547,
      "eval_rewards/rejected": -0.833840548992157,
      "eval_runtime": 68.9632,
      "eval_samples_per_second": 1.45,
      "eval_steps_per_second": 0.363,
      "step": 10
    },
    {
      "epoch": 0.32,
      "eval_logits/chosen": -2.7328505516052246,
      "eval_logits/rejected": -3.097344398498535,
      "eval_logps/chosen": -248.98631286621094,
      "eval_logps/rejected": -93.10418701171875,
      "eval_loss": 1.2917519143229583e-07,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 14.851668357849121,
      "eval_rewards/margins": 17.661312103271484,
      "eval_rewards/rejected": -2.8096444606781006,
      "eval_runtime": 68.7434,
      "eval_samples_per_second": 1.455,
      "eval_steps_per_second": 0.364,
      "step": 20
    },
    {
      "epoch": 0.48,
      "eval_logits/chosen": -2.710421562194824,
      "eval_logits/rejected": -3.082587957382202,
      "eval_logps/chosen": -246.4652862548828,
      "eval_logps/rejected": -98.84378814697266,
      "eval_loss": 7.008074476289039e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 15.103771209716797,
      "eval_rewards/margins": 18.487375259399414,
      "eval_rewards/rejected": -3.3836045265197754,
      "eval_runtime": 68.5884,
      "eval_samples_per_second": 1.458,
      "eval_steps_per_second": 0.364,
      "step": 30
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.7037651538848877,
      "eval_logits/rejected": -3.078204870223999,
      "eval_logps/chosen": -246.36607360839844,
      "eval_logps/rejected": -100.4520492553711,
      "eval_loss": 6.257658924369025e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 15.113690376281738,
      "eval_rewards/margins": 18.658119201660156,
      "eval_rewards/rejected": -3.5444304943084717,
      "eval_runtime": 68.5134,
      "eval_samples_per_second": 1.46,
      "eval_steps_per_second": 0.365,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.187412519560894e-06,
      "learning_rate": 2e-05,
      "logits/chosen": -2.744260311126709,
      "logits/rejected": -3.1097443103790283,
      "logps/chosen": -273.5040283203125,
      "logps/rejected": -87.8642807006836,
      "loss": 0.0898,
      "rewards/accuracies": 0.9474999904632568,
      "rewards/chosen": 12.438501358032227,
      "rewards/margins": 14.762829780578613,
      "rewards/rejected": -2.3243308067321777,
      "step": 50
    },
    {
      "epoch": 0.8,
      "eval_logits/chosen": -2.7021143436431885,
      "eval_logits/rejected": -3.077101945877075,
      "eval_logps/chosen": -246.37890625,
      "eval_logps/rejected": -100.837890625,
      "eval_loss": 6.107774197516846e-08,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 15.112407684326172,
      "eval_rewards/margins": 18.695423126220703,
      "eval_rewards/rejected": -3.583014488220215,
      "eval_runtime": 68.5614,
      "eval_samples_per_second": 1.459,
      "eval_steps_per_second": 0.365,
      "step": 50
    }
  ],
  "logging_steps": 50,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "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": 4,
  "trial_name": null,
  "trial_params": null
}