{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0021746221593998043,
  "eval_steps": 500,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005436555398499511,
      "grad_norm": 14.280742645263672,
      "learning_rate": 8.571428571428571e-07,
      "logits/chosen": 0.7501829266548157,
      "logits/rejected": 0.7856999635696411,
      "logps/chosen": -277.14459228515625,
      "logps/rejected": -273.17779541015625,
      "loss": 0.6951,
      "rewards/accuracies": 0.25,
      "rewards/chosen": -0.0013596273493021727,
      "rewards/margins": -0.003496611025184393,
      "rewards/rejected": 0.0021369843743741512,
      "step": 5
    },
    {
      "epoch": 0.0010873110796999022,
      "grad_norm": 12.856571197509766,
      "learning_rate": 1.9285714285714285e-06,
      "logits/chosen": 0.6582454442977905,
      "logits/rejected": 0.6293224096298218,
      "logps/chosen": -274.37469482421875,
      "logps/rejected": -226.21829223632812,
      "loss": 0.6888,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.006261076778173447,
      "rewards/margins": 0.009591618552803993,
      "rewards/rejected": -0.0033305410761386156,
      "step": 10
    },
    {
      "epoch": 0.0016309666195498532,
      "grad_norm": 12.676877975463867,
      "learning_rate": 3e-06,
      "logits/chosen": 0.7257691621780396,
      "logits/rejected": 0.6618317365646362,
      "logps/chosen": -273.8007507324219,
      "logps/rejected": -254.3082733154297,
      "loss": 0.6888,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.011636307463049889,
      "rewards/margins": 0.009717724286019802,
      "rewards/rejected": 0.0019185834098607302,
      "step": 15
    },
    {
      "epoch": 0.0021746221593998043,
      "grad_norm": 10.399931907653809,
      "learning_rate": 4.071428571428571e-06,
      "logits/chosen": 0.737468957901001,
      "logits/rejected": 0.7028112411499023,
      "logps/chosen": -273.6678466796875,
      "logps/rejected": -292.0629577636719,
      "loss": 0.6898,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": 0.011708216741681099,
      "rewards/margins": 0.008118860423564911,
      "rewards/rejected": 0.0035893558524549007,
      "step": 20
    },
    {
      "epoch": 0.0021746221593998043,
      "eval_logits/chosen": 0.6203805804252625,
      "eval_logits/rejected": 0.4696192145347595,
      "eval_logps/chosen": -275.54425048828125,
      "eval_logps/rejected": -254.13768005371094,
      "eval_loss": 0.6896114945411682,
      "eval_rewards/accuracies": 0.5400000214576721,
      "eval_rewards/chosen": 0.022015957161784172,
      "eval_rewards/margins": 0.00875764712691307,
      "eval_rewards/rejected": 0.013258310966193676,
      "eval_runtime": 35.197,
      "eval_samples_per_second": 5.682,
      "eval_steps_per_second": 5.682,
      "step": 20
    }
  ],
  "logging_steps": 5,
  "max_steps": 27591,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200.0,
  "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": 8,
  "trial_name": null,
  "trial_params": null
}