all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": -2.228393316268921,
    "eval_logits/rejected": -2.186522960662842,
    "eval_logps/chosen": -110.88995361328125,
    "eval_logps/rejected": -120.80963897705078,
    "eval_loss": 0.5066802501678467,
    "eval_rewards/accuracies": 0.7098214030265808,
    "eval_rewards/chosen": -1.1023340225219727,
    "eval_rewards/margins": 1.2738500833511353,
    "eval_rewards/rejected": -2.3761842250823975,
    "eval_runtime": 487.8244,
    "eval_samples": 1786,
    "eval_samples_per_second": 3.661,
    "eval_steps_per_second": 0.115,
    "train_loss": 0.58384587518933,
    "train_runtime": 30698.0699,
    "train_samples": 53693,
    "train_samples_per_second": 1.749,
    "train_steps_per_second": 0.109
}