all_results.json

{
    "epoch": 4.988190836088805,
    "eval_dpo_loss": 0.7149785757064819,
    "eval_logits": -0.6811044812202454,
    "eval_logps": -80.53881072998047,
    "eval_loss": 0.5835002660751343,
    "eval_objective": 0.5760073065757751,
    "eval_ranking_idealized": 0.5888429880142212,
    "eval_ranking_idealized_expo": 0.5103305578231812,
    "eval_ranking_simple": 0.5247933864593506,
    "eval_regularize": 0.5760073065757751,
    "eval_runtime": 308.7188,
    "eval_samples": 5790,
    "eval_samples_per_second": 18.755,
    "eval_steps_per_second": 0.784,
    "total_flos": 0.0,
    "train_loss": 0.1997517315332185,
    "train_runtime": 41462.5211,
    "train_samples": 50802,
    "train_samples_per_second": 6.126,
    "train_steps_per_second": 0.021
}