all_results.json

{
    "epoch": 1.0,
    "eval_logits/chosen": 0.17750975489616394,
    "eval_logits/rejected": 0.2212180495262146,
    "eval_logps/chosen": -133.01571655273438,
    "eval_logps/rejected": -145.18185424804688,
    "eval_loss": 1.3258469104766846,
    "eval_nll_loss": 0.3315570652484894,
    "eval_rewards/accuracies": 0.7177419066429138,
    "eval_rewards/chosen": -13.3015718460083,
    "eval_rewards/margins": 1.216615080833435,
    "eval_rewards/rejected": -14.518187522888184,
    "eval_runtime": 102.7641,
    "eval_samples": 1961,
    "eval_samples_per_second": 19.083,
    "eval_steps_per_second": 0.302,
    "total_flos": 0.0,
    "train_loss": 1.4382328207676227,
    "train_runtime": 9608.0854,
    "train_samples": 59876,
    "train_samples_per_second": 6.232,
    "train_steps_per_second": 0.049
}