all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.945185899734497,
    "eval_logits/rejected": -2.9309117794036865,
    "eval_logps/chosen": -39.84721755981445,
    "eval_logps/rejected": -1156.61669921875,
    "eval_loss": 2.3733691705274396e-05,
    "eval_rewards/accuracies": 1.0,
    "eval_rewards/chosen": 0.1952119767665863,
    "eval_rewards/margins": 11.378995895385742,
    "eval_rewards/rejected": -11.183783531188965,
    "eval_runtime": 3.9067,
    "eval_samples": 5,
    "eval_samples_per_second": 1.28,
    "eval_steps_per_second": 0.256,
    "train_loss": 0.02337361431049117,
    "train_runtime": 68623.0806,
    "train_samples": 134157,
    "train_samples_per_second": 1.955,
    "train_steps_per_second": 0.244
}