all_results.json

{
    "epoch": 0.9999375663357682,
    "eval_logits/chosen": 0.6540641784667969,
    "eval_logits/rejected": 1.569779634475708,
    "eval_logps/chosen": -84.51407623291016,
    "eval_logps/rejected": -994.2071533203125,
    "eval_loss": 0.00282670627348125,
    "eval_rewards/accuracies": 1.0,
    "eval_rewards/chosen": -0.40158194303512573,
    "eval_rewards/margins": 8.929591178894043,
    "eval_rewards/rejected": -9.33117389678955,
    "eval_runtime": 0.6352,
    "eval_samples": 5,
    "eval_samples_per_second": 7.871,
    "eval_steps_per_second": 7.871,
    "total_flos": 0.0,
    "train_loss": 0.05122942747121405,
    "train_runtime": 6577.5594,
    "train_samples": 16017,
    "train_samples_per_second": 2.435,
    "train_steps_per_second": 0.609
}