all_results.json

{
    "epoch": 9.995276334435522,
    "eval_dpo_loss": 0.708280622959137,
    "eval_logits": -1.1884795427322388,
    "eval_logps": -93.80847930908203,
    "eval_loss": 0.3417203724384308,
    "eval_objective": 0.34029480814933777,
    "eval_ranking_idealized": 0.5398550629615784,
    "eval_ranking_idealized_expo": 0.5243270993232727,
    "eval_ranking_simple": 0.5186335444450378,
    "eval_regularize": 0.26946672797203064,
    "eval_runtime": 154.888,
    "eval_samples": 5790,
    "eval_samples_per_second": 37.382,
    "eval_steps_per_second": 3.118,
    "total_flos": 0.0,
    "train_loss": 0.15836801591183153,
    "train_runtime": 49865.2618,
    "train_samples": 50802,
    "train_samples_per_second": 10.188,
    "train_steps_per_second": 0.106
}