eval_results.json

{
    "epoch": 1.8422295701464337,
    "eval_dpo_loss": 2.573197364807129,
    "eval_logits": -0.484580934047699,
    "eval_logps": -80.18506622314453,
    "eval_loss": 4.899093151092529,
    "eval_objective": 4.920805931091309,
    "eval_ranking_idealized": 0.5295031070709229,
    "eval_ranking_idealized_expo": 0.5212215185165405,
    "eval_ranking_simple": 0.523809552192688,
    "eval_regularize": 4.920805931091309,
    "eval_runtime": 307.8846,
    "eval_samples": 5790,
    "eval_samples_per_second": 18.806,
    "eval_steps_per_second": 1.569,
    "eval_wo_beta": 14.127137184143066
}