{
    "epoch": 4.988190836088805,
    "eval_dpo_loss": 0.6860232353210449,
    "eval_logits": -0.8985373377799988,
    "eval_logps": -77.20108032226562,
    "eval_loss": 0.4401702880859375,
    "eval_objective": 0.4385421574115753,
    "eval_ranking_idealized": 0.6570248007774353,
    "eval_ranking_idealized_expo": 0.5113636255264282,
    "eval_ranking_simple": 0.5320248007774353,
    "eval_regularize": 0.4385421574115753,
    "eval_runtime": 259.9589,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.273,
    "eval_steps_per_second": 0.931,
    "total_flos": 0.0,
    "train_loss": 0.15004684163088147,
    "train_runtime": 35354.4588,
    "train_samples": 50802,
    "train_samples_per_second": 7.185,
    "train_steps_per_second": 0.025
}