{ "epoch": 0.9990186457311089, "eval_logits/chosen": -1.1953125, "eval_logits/rejected": -1.2421875, "eval_logps/chosen": -446.0, "eval_logps/chosen_bottom_tokens": -14.375, "eval_logps/chosen_top_tokens": -0.00072479248046875, "eval_logps/rejected": -416.0, "eval_logps/rejected_bottom_tokens": -14.3125, "eval_logps/rejected_top_tokens": -0.000728607177734375, "eval_loss": 0.6256738305091858, "eval_rewards/accuracies": 0.6597015857696533, "eval_rewards/chosen": -0.5234375, "eval_rewards/margins": 0.2578125, "eval_rewards/rejected": -0.78125, "eval_runtime": 111.3149, "eval_samples": 2000, "eval_samples_per_second": 17.967, "eval_steps_per_second": 0.602 }