{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 48,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 1e-07,
      "logits/chosen": -2.732227325439453,
      "logits/rejected": -2.681589126586914,
      "logps/chosen": -155.3662109375,
      "logps/pi_response": -102.39925384521484,
      "logps/ref_response": -102.39925384521484,
      "logps/rejected": -336.2640380859375,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.83504027183137e-07,
      "logits/chosen": -2.641280174255371,
      "logits/rejected": -2.598306894302368,
      "logps/chosen": -277.44879150390625,
      "logps/pi_response": -145.1143341064453,
      "logps/ref_response": -140.9521484375,
      "logps/rejected": -374.81488037109375,
      "loss": 0.66,
      "rewards/accuracies": 0.5555555820465088,
      "rewards/chosen": -0.14305411279201508,
      "rewards/margins": 0.12196901440620422,
      "rewards/rejected": -0.2650231420993805,
      "step": 10
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.643105808261596e-07,
      "logits/chosen": -2.619703531265259,
      "logits/rejected": -2.5993566513061523,
      "logps/chosen": -301.9560241699219,
      "logps/pi_response": -140.05838012695312,
      "logps/ref_response": -132.67494201660156,
      "logps/rejected": -493.3030700683594,
      "loss": 0.6149,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": -0.627299427986145,
      "rewards/margins": 0.8519609570503235,
      "rewards/rejected": -1.4792603254318237,
      "step": 20
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8676665440207977e-07,
      "logits/chosen": -2.6581668853759766,
      "logits/rejected": -2.59602689743042,
      "logps/chosen": -296.7784423828125,
      "logps/pi_response": -136.73129272460938,
      "logps/ref_response": -131.84959411621094,
      "logps/rejected": -463.1182556152344,
      "loss": 0.5468,
      "rewards/accuracies": 0.703125,
      "rewards/chosen": -0.5024408102035522,
      "rewards/margins": 0.6750184893608093,
      "rewards/rejected": -1.1774593591690063,
      "step": 30
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1500545527530544e-08,
      "logits/chosen": -2.6679675579071045,
      "logits/rejected": -2.6350443363189697,
      "logps/chosen": -290.5754699707031,
      "logps/pi_response": -128.8948974609375,
      "logps/ref_response": -123.4980697631836,
      "logps/rejected": -458.8999938964844,
      "loss": 0.5293,
      "rewards/accuracies": 0.7406250238418579,
      "rewards/chosen": -0.4711049497127533,
      "rewards/margins": 0.6303536295890808,
      "rewards/rejected": -1.1014585494995117,
      "step": 40
    },
    {
      "epoch": 1.0,
      "step": 48,
      "total_flos": 0.0,
      "train_loss": 0.5820941229661306,
      "train_runtime": 2881.4015,
      "train_samples_per_second": 4.243,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 10,
  "max_steps": 48,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}