{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005211431160959583,
  "eval_steps": 400,
  "global_step": 23,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001338016390700786,
      "grad_norm": 1640.0,
      "learning_rate": 7.464101615137755e-08,
      "logits/chosen": -0.8070744276046753,
      "logits/rejected": -0.6946424841880798,
      "logps/chosen": -1.7985578775405884,
      "logps/rejected": -1.9126713275909424,
      "loss": 82.739,
      "rewards/accuracies": 0.512499988079071,
      "rewards/chosen": -17.985576629638672,
      "rewards/margins": 1.141135334968567,
      "rewards/rejected": -19.126712799072266,
      "step": 5
    },
    {
      "epoch": 0.002676032781401572,
      "grad_norm": 1448.0,
      "learning_rate": 4.6945927106677224e-08,
      "logits/chosen": -0.6592462658882141,
      "logits/rejected": -0.5781775712966919,
      "logps/chosen": -1.62754225730896,
      "logps/rejected": -1.859715461730957,
      "loss": 82.6139,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": -16.27542495727539,
      "rewards/margins": 2.3217310905456543,
      "rewards/rejected": -18.59715461730957,
      "step": 10
    },
    {
      "epoch": 0.004014049172102358,
      "grad_norm": 2352.0,
      "learning_rate": 1.4288495612538426e-08,
      "logits/chosen": -0.7022191882133484,
      "logits/rejected": -0.6638501286506653,
      "logps/chosen": -1.8131605386734009,
      "logps/rejected": -1.7845131158828735,
      "loss": 106.2027,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -18.131603240966797,
      "rewards/margins": -0.2864739000797272,
      "rewards/rejected": -17.84513282775879,
      "step": 15
    },
    {
      "epoch": 0.005352065562803144,
      "grad_norm": 2496.0,
      "learning_rate": 0.0,
      "logits/chosen": -0.6888337135314941,
      "logits/rejected": -0.650062084197998,
      "logps/chosen": -1.786657691001892,
      "logps/rejected": -1.9025501012802124,
      "loss": 96.0509,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -17.8665771484375,
      "rewards/margins": 1.1589242219924927,
      "rewards/rejected": -19.025501251220703,
      "step": 20
    },
    {
      "epoch": 0.005211431160959583,
      "step": 23,
      "total_flos": 0.0,
      "train_loss": 3.4047957710597827,
      "train_runtime": 14.6858,
      "train_samples_per_second": 21.79,
      "train_steps_per_second": 1.362
    }
  ],
  "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": 1,
  "trial_name": null,
  "trial_params": null
}