{
    "epoch": 1.0,
    "eval_logits/chosen": -2.265434980392456,
    "eval_logits/rejected": -2.236332654953003,
    "eval_logps/chosen": -312.9822692871094,
    "eval_logps/rejected": -267.5411682128906,
    "eval_loss": 0.6714204549789429,
    "eval_rewards/accuracies": 0.6875,
    "eval_rewards/chosen": 0.1837889701128006,
    "eval_rewards/margins": 0.15236972272396088,
    "eval_rewards/rejected": 0.031419236212968826,
    "eval_runtime": 72.2288,
    "eval_samples": 256,
    "eval_samples_per_second": 3.544,
    "eval_steps_per_second": 0.055,
    "train_loss": 0.6752520862378573,
    "train_runtime": 2732.0836,
    "train_samples": 4860,
    "train_samples_per_second": 1.779,
    "train_steps_per_second": 0.028
}