{
    "epoch": 1.0,
    "eval_kl/n_epsilon_steps": 0.3836805522441864,
    "eval_kl/p_epsilon_steps": 0.6158854365348816,
    "eval_logits/chosen": -0.8295226693153381,
    "eval_logits/rejected": -0.7224703431129456,
    "eval_logps/chosen": -208.40182495117188,
    "eval_logps/ref_chosen": -87.82356262207031,
    "eval_logps/ref_rejected": -82.81887817382812,
    "eval_logps/rejected": -244.0940704345703,
    "eval_loss": 0.6479349136352539,
    "eval_rewards/accuracies": 0.6414930820465088,
    "eval_rewards/chosen": -0.3830554485321045,
    "eval_rewards/margins": 0.12642447650432587,
    "eval_rewards/rejected": -0.5094798803329468,
    "eval_runtime": 22.2873,
    "eval_samples": 2339,
    "eval_samples_per_second": 104.948,
    "eval_steps_per_second": 0.853,
    "total_flos": 0.0,
    "train_loss": 0.6232832217917723,
    "train_runtime": 1489.7896,
    "train_samples": 43598,
    "train_samples_per_second": 29.265,
    "train_steps_per_second": 0.228
}