all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.842418670654297,
    "eval_logits/rejected": -2.846235752105713,
    "eval_logps/chosen": -284.122314453125,
    "eval_logps/rejected": -259.4594421386719,
    "eval_loss": 0.6635700464248657,
    "eval_pred_label": 4797.50390625,
    "eval_rewards/accuracies": 0.628000020980835,
    "eval_rewards/chosen": 0.013506044633686543,
    "eval_rewards/margins": 0.029479000717401505,
    "eval_rewards/rejected": -0.015972958877682686,
    "eval_runtime": 439.3222,
    "eval_samples": 2000,
    "eval_samples_per_second": 4.552,
    "eval_steps_per_second": 0.285,
    "eval_use_label": 11234.49609375,
    "train_loss": 0.6728555943953429,
    "train_runtime": 24272.064,
    "train_samples": 61135,
    "train_samples_per_second": 2.519,
    "train_steps_per_second": 0.039
}