{
    "epoch": 1.0,
    "eval_logits/chosen": 4.460368633270264,
    "eval_logits/rejected": 4.594798564910889,
    "eval_logps/chosen": -380.465087890625,
    "eval_logps/rejected": -579.1896362304688,
    "eval_loss": 0.4399436116218567,
    "eval_pred_label": 2212.28125,
    "eval_rewards/accuracies": 0.33984375,
    "eval_rewards/chosen": -3.1655280590057373,
    "eval_rewards/margins": 1.8544387817382812,
    "eval_rewards/rejected": -5.019967079162598,
    "eval_runtime": 125.2743,
    "eval_samples": 2000,
    "eval_samples_per_second": 15.965,
    "eval_steps_per_second": 0.255,
    "eval_use_label": 6575.71875
}