{
    "epoch": 2.975909305621162,
    "eval_dpo_loss": 0.6790949106216431,
    "eval_logits": -1.383711576461792,
    "eval_logps": -89.10604858398438,
    "eval_loss": 0.4217349588871002,
    "eval_objective": 0.4141845107078552,
    "eval_ranking_idealized": 0.6030020713806152,
    "eval_ranking_idealized_expo": 0.5222567319869995,
    "eval_ranking_simple": 0.534679114818573,
    "eval_regularize": 0.4141845107078552,
    "eval_runtime": 467.1821,
    "eval_samples": 5790,
    "eval_samples_per_second": 12.393,
    "eval_steps_per_second": 1.034,
    "eval_wo_beta": 15.984663963317871,
    "total_flos": 0.0,
    "train_loss": 0.10327800432840983,
    "train_runtime": 29791.2438,
    "train_samples": 50802,
    "train_samples_per_second": 8.526,
    "train_steps_per_second": 0.059
}