eval_results.json

{
    "epoch": 4.988190836088805,
    "eval_dpo_loss": 0.6829335689544678,
    "eval_logits": -1.4302784204483032,
    "eval_logps": -90.04464721679688,
    "eval_loss": 0.40619462728500366,
    "eval_objective": 0.4077059328556061,
    "eval_ranking_idealized": 0.5888429880142212,
    "eval_ranking_idealized_expo": 0.5103305578231812,
    "eval_ranking_simple": 0.5247933864593506,
    "eval_regularize": 0.4077059328556061,
    "eval_runtime": 259.5368,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.309,
    "eval_steps_per_second": 0.932
}