all_results.json

{
    "epoch": 0.0032112393376818866,
    "eval_logits/chosen": -0.35490095615386963,
    "eval_logits/rejected": -0.32419130206108093,
    "eval_logps/chosen": -1.7861407995224,
    "eval_logps/rejected": -1.8718520402908325,
    "eval_loss": 6.484124660491943,
    "eval_rewards/accuracies": 0.5215133428573608,
    "eval_rewards/chosen": -17.861406326293945,
    "eval_rewards/margins": 0.8571121692657471,
    "eval_rewards/rejected": -18.71851921081543,
    "eval_runtime": 59.8932,
    "eval_samples": 1345,
    "eval_samples_per_second": 22.457,
    "eval_steps_per_second": 5.627,
    "total_flos": 0.0,
    "train_loss": 0.40756722291310626,
    "train_runtime": 13.7037,
    "train_samples": 59790,
    "train_samples_per_second": 11.676,
    "train_steps_per_second": 0.73
}