all_results.json

{
    "epoch": 3.119981105337742,
    "eval_dpo_loss": 0.7208675742149353,
    "eval_logits": -0.5211770534515381,
    "eval_logps": -81.06893157958984,
    "eval_loss": 283.0077819824219,
    "eval_objective": 277.3703308105469,
    "eval_ranking_idealized": 0.6030020713806152,
    "eval_ranking_idealized_expo": 0.5222567319869995,
    "eval_ranking_simple": 0.5331262946128845,
    "eval_regularize": 0.6309527158737183,
    "eval_runtime": 308.4993,
    "eval_samples": 5790,
    "eval_samples_per_second": 18.768,
    "eval_steps_per_second": 1.566,
    "eval_wo_beta": 14.269521713256836,
    "total_flos": 0.0,
    "train_loss": 50.30704811789773,
    "train_runtime": 15116.5529,
    "train_samples": 50802,
    "train_samples_per_second": 16.803,
    "train_steps_per_second": 0.116
}