{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.8173785209655762,
      "learning_rate": 7.435897435897435e-07,
      "logits/chosen": 0.13414913415908813,
      "logits/rejected": 0.12645891308784485,
      "logps/chosen": -298.76983642578125,
      "logps/rejected": -261.02435302734375,
      "loss": 9.7803,
      "nll_loss": 1.0939728021621704,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -29.87698745727539,
      "rewards/margins": -3.774548292160034,
      "rewards/rejected": -26.102436065673828,
      "step": 10
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 1.8889318704605103,
      "learning_rate": 4.871794871794871e-07,
      "logits/chosen": 0.077309250831604,
      "logits/rejected": 0.12109130620956421,
      "logps/chosen": -277.27899169921875,
      "logps/rejected": -260.14141845703125,
      "loss": 8.6446,
      "nll_loss": 1.0915288925170898,
      "rewards/accuracies": 0.4749999940395355,
      "rewards/chosen": -27.7278995513916,
      "rewards/margins": -1.7137558460235596,
      "rewards/rejected": -26.014141082763672,
      "step": 20
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 1.609092354774475,
      "learning_rate": 2.3076923076923078e-07,
      "logits/chosen": 0.09378460794687271,
      "logits/rejected": 0.0867479220032692,
      "logps/chosen": -280.64697265625,
      "logps/rejected": -273.1399841308594,
      "loss": 7.8556,
      "nll_loss": 1.1089527606964111,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -28.064701080322266,
      "rewards/margins": -0.7507012486457825,
      "rewards/rejected": -27.31399917602539,
      "step": 30
    },
    {
      "epoch": 3.0,
      "step": 39,
      "total_flos": 0.0,
      "train_loss": 8.851261627979767,
      "train_runtime": 435.0584,
      "train_samples_per_second": 0.69,
      "train_steps_per_second": 0.09
    }
  ],
  "logging_steps": 10,
  "max_steps": 39,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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
}