{ "epoch": 1.0, "eval_dpo_losses": 0.6843776702880859, "eval_logits/chosen": -2.7839062213897705, "eval_logits/rejected": -2.7455294132232666, "eval_logps/chosen": -279.266357421875, "eval_logps/rejected": -255.0995635986328, "eval_loss": 0.7023824453353882, "eval_positive_losses": 0.16172000765800476, "eval_rewards/accuracies": 0.593999981880188, "eval_rewards/chosen": 0.05327073112130165, "eval_rewards/margins": 0.018477285280823708, "eval_rewards/margins_max": 0.10823574662208557, "eval_rewards/margins_min": -0.059972915798425674, "eval_rewards/margins_std": 0.05609079822897911, "eval_rewards/rejected": 0.034793443977832794, "eval_runtime": 427.9806, "eval_samples": 2000, "eval_samples_per_second": 4.673, "eval_steps_per_second": 0.292 }