{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 100,
  "global_step": 45,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "grad_norm": 14.650781993147133,
      "learning_rate": 1e-07,
      "logits/chosen": -1.8544178009033203,
      "logits/rejected": -1.5131595134735107,
      "logps/chosen": -382.39324951171875,
      "logps/rejected": -74.34650421142578,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.67,
      "grad_norm": 13.355694493789834,
      "learning_rate": 4.809698831278217e-07,
      "logits/chosen": -1.7981030941009521,
      "logits/rejected": -1.8637254238128662,
      "logps/chosen": -297.70697021484375,
      "logps/rejected": -91.71724700927734,
      "loss": 0.6887,
      "rewards/accuracies": 0.625,
      "rewards/chosen": 0.0073351990431547165,
      "rewards/margins": 0.007683979347348213,
      "rewards/rejected": -0.000348779110936448,
      "step": 10
    },
    {
      "epoch": 1.33,
      "grad_norm": 10.957160162463524,
      "learning_rate": 3.4567085809127245e-07,
      "logits/chosen": -1.7605440616607666,
      "logits/rejected": -1.6980727910995483,
      "logps/chosen": -317.1847229003906,
      "logps/rejected": -89.17781066894531,
      "loss": 0.6424,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": 0.10211040079593658,
      "rewards/margins": 0.10564006865024567,
      "rewards/rejected": -0.003529661800712347,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.631401952273302,
      "learning_rate": 1.5432914190872756e-07,
      "logits/chosen": -1.7424694299697876,
      "logits/rejected": -1.690342903137207,
      "logps/chosen": -304.099609375,
      "logps/rejected": -91.2957763671875,
      "loss": 0.5679,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.263149231672287,
      "rewards/margins": 0.2803087830543518,
      "rewards/rejected": -0.017159538343548775,
      "step": 30
    },
    {
      "epoch": 2.67,
      "grad_norm": 9.39281581673066,
      "learning_rate": 1.9030116872178314e-08,
      "logits/chosen": -1.7320054769515991,
      "logits/rejected": -1.6545593738555908,
      "logps/chosen": -303.2212829589844,
      "logps/rejected": -99.20294189453125,
      "loss": 0.5282,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.36085695028305054,
      "rewards/margins": 0.40523195266723633,
      "rewards/rejected": -0.04437502473592758,
      "step": 40
    },
    {
      "epoch": 3.0,
      "step": 45,
      "total_flos": 0.0,
      "train_loss": 0.5981406688690185,
      "train_runtime": 346.5059,
      "train_samples_per_second": 16.32,
      "train_steps_per_second": 0.13
    }
  ],
  "logging_steps": 10,
  "max_steps": 45,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}