{
    "epoch": 2.9984268484530676,
    "eval_log_odds_chosen": 1.220078706741333,
    "eval_log_odds_ratio": -1.047989010810852,
    "eval_logits/chosen": -2.849764585494995,
    "eval_logits/rejected": -2.8669533729553223,
    "eval_logps/chosen": -2.7812891006469727,
    "eval_logps/rejected": -3.8601787090301514,
    "eval_loss": 1.3770909309387207,
    "eval_nll_loss": 1.3531930446624756,
    "eval_rewards/accuracies": 0.6527777910232544,
    "eval_rewards/chosen": -0.13906444609165192,
    "eval_rewards/margins": 0.05394447594881058,
    "eval_rewards/rejected": -0.1930089294910431,
    "eval_runtime": 135.9849,
    "eval_samples": 1994,
    "eval_samples_per_second": 14.663,
    "eval_steps_per_second": 0.463,
    "total_flos": 0.0,
    "train_loss": 0.32389816019492534,
    "train_runtime": 62235.4926,
    "train_samples": 61005,
    "train_samples_per_second": 2.941,
    "train_steps_per_second": 0.046
}