all_results.json

{
    "epoch": 1.995276334435522,
    "eval_dpo_loss": 0.6890939474105835,
    "eval_logits": -1.5636364221572876,
    "eval_logps": -92.05500793457031,
    "eval_loss": 0.6817956566810608,
    "eval_objective": 0.6890939474105835,
    "eval_ranking_idealized": 0.5888429880142212,
    "eval_ranking_idealized_expo": 0.5103305578231812,
    "eval_ranking_simple": 0.51962810754776,
    "eval_regularize": 0.6890939474105835,
    "eval_runtime": 259.6459,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.3,
    "eval_steps_per_second": 0.932,
    "total_flos": 0.0,
    "train_loss": 0.6593505936590108,
    "train_runtime": 15975.3291,
    "train_samples": 50802,
    "train_samples_per_second": 6.36,
    "train_steps_per_second": 0.022
}