{
    "epoch": 3.0,
    "eval_dpo_losses": 0.6657702922821045,
    "eval_logits/chosen": -2.7033236026763916,
    "eval_logits/rejected": -2.660062789916992,
    "eval_logps/chosen": -289.4331359863281,
    "eval_logps/rejected": -270.23931884765625,
    "eval_loss": 1.5678311586380005,
    "eval_positive_losses": 8.853089332580566,
    "eval_rewards/accuracies": 0.6150793433189392,
    "eval_rewards/chosen": -0.04211931675672531,
    "eval_rewards/margins": 0.0684470534324646,
    "eval_rewards/margins_max": 0.31783527135849,
    "eval_rewards/margins_min": -0.1569632887840271,
    "eval_rewards/margins_std": 0.2114361971616745,
    "eval_rewards/rejected": -0.11056637018918991,
    "eval_runtime": 345.2654,
    "eval_samples": 2000,
    "eval_samples_per_second": 5.793,
    "eval_steps_per_second": 0.182,
    "train_loss": 0.5596254680078354,
    "train_runtime": 11165.5936,
    "train_samples": 5678,
    "train_samples_per_second": 1.526,
    "train_steps_per_second": 0.095
}