{
    "epoch": 1.0,
    "eval_logits/chosen": -2.9375,
    "eval_logits/rejected": -3.078125,
    "eval_logps/chosen": -564.0,
    "eval_logps/rejected": -900.0,
    "eval_loss": 0.2485724240541458,
    "eval_rewards/accuracies": 0.8979707956314087,
    "eval_rewards/chosen": -3.65625,
    "eval_rewards/margins": 3.59375,
    "eval_rewards/rejected": -7.25,
    "eval_runtime": 2905.6531,
    "eval_samples": 99217,
    "eval_samples_per_second": 33.9,
    "eval_steps_per_second": 0.53,
    "total_flos": 0.0,
    "train_loss": 0.23414572194140112,
    "train_runtime": 130626.0521,
    "train_samples": 1773152,
    "train_samples_per_second": 13.574,
    "train_steps_per_second": 0.106
}