{
    "epoch": 2.0,
    "eval_logits/chosen": -2.668743848800659,
    "eval_logits/rejected": -2.614588499069214,
    "eval_logps/chosen": -48.75492858886719,
    "eval_logps/rejected": -238.03585815429688,
    "eval_loss": 0.010777823626995087,
    "eval_rewards/accuracies": 0.9956896305084229,
    "eval_rewards/chosen": -1.6998504400253296,
    "eval_rewards/margins": 20.81882667541504,
    "eval_rewards/rejected": -22.5186767578125,
    "eval_runtime": 231.3578,
    "eval_samples": 905,
    "eval_samples_per_second": 3.912,
    "eval_steps_per_second": 0.125,
    "train_loss": 0.06552632141962543,
    "train_runtime": 3687.7187,
    "train_samples": 4570,
    "train_samples_per_second": 2.478,
    "train_steps_per_second": 0.078
}