all_results.json

{
    "epoch": 1.995276334435522,
    "eval_dpo_loss": 1520.4888916015625,
    "eval_logits": -1.0707257986068726,
    "eval_logps": -79.1167984008789,
    "eval_loss": 1521.187255859375,
    "eval_objective": 1520.4888916015625,
    "eval_ranking_idealized": 0.5092975497245789,
    "eval_ranking_idealized_expo": 0.5092975497245789,
    "eval_ranking_simple": 0.5258264541625977,
    "eval_regularize": 1520.4888916015625,
    "eval_runtime": 261.1563,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.171,
    "eval_steps_per_second": 0.927,
    "total_flos": 0.0,
    "train_loss": 584.4622564356436,
    "train_runtime": 14141.3958,
    "train_samples": 50802,
    "train_samples_per_second": 7.185,
    "train_steps_per_second": 0.025
}