eval_results.json

{
    "epoch": 3.4057628719886632,
    "eval_dpo_loss": 26.279430389404297,
    "eval_logits": -0.5139885544776917,
    "eval_logps": -80.61864471435547,
    "eval_loss": 51.679359436035156,
    "eval_objective": 51.56280517578125,
    "eval_ranking_idealized": 0.5212215185165405,
    "eval_ranking_idealized_expo": 0.5212215185165405,
    "eval_ranking_simple": 0.5248447060585022,
    "eval_regularize": 51.56280517578125,
    "eval_runtime": 307.6436,
    "eval_samples": 5790,
    "eval_samples_per_second": 18.82,
    "eval_steps_per_second": 1.57,
    "eval_wo_beta": 14.074385643005371
}