eval_results.json

{
    "epoch": 4.988190836088805,
    "eval_dpo_loss": 0.7514079809188843,
    "eval_logits": -1.2688928842544556,
    "eval_logps": -87.71137237548828,
    "eval_loss": 0.7094846963882446,
    "eval_objective": 0.7116978764533997,
    "eval_ranking_idealized": 0.5247933864593506,
    "eval_ranking_idealized_expo": 0.5092975497245789,
    "eval_ranking_simple": 0.5123966932296753,
    "eval_regularize": 0.7116978764533997,
    "eval_runtime": 259.375,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.323,
    "eval_steps_per_second": 0.933
}