all_results.json

{
    "epoch": 1.995276334435522,
    "eval_dpo_loss": 1.605820894241333,
    "eval_logits": -1.134599208831787,
    "eval_logps": -81.46167755126953,
    "eval_loss": 1.5849758386611938,
    "eval_objective": 1.605820894241333,
    "eval_ranking_idealized": 0.5144628286361694,
    "eval_ranking_idealized_expo": 0.5092975497245789,
    "eval_ranking_simple": 0.5216942429542542,
    "eval_regularize": 1.605820894241333,
    "eval_runtime": 259.039,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.352,
    "eval_steps_per_second": 0.934,
    "total_flos": 0.0,
    "train_loss": 0.635454860092564,
    "train_runtime": 14182.6927,
    "train_samples": 50802,
    "train_samples_per_second": 7.164,
    "train_steps_per_second": 0.025
}