In this paper, we identify and characterize the emerging area of representation engineering (RepE), an approach to enhancing the transparency of AI systems that draws on insights from cognitive neuroscience. RepE places population-level representations, rather than neurons or circuits, at the center of analysis, equipping us with novel methods for monitoring and manipulating high-level cognitive phenomena in deep neural networks (DNNs). We provide baselines and an initial analysis of RepE techniques, showing that they offer simple yet effective solutions for improving our understanding and control of large language models. We showcase how these methods can provide traction on a wide range of safety-relevant problems, including honesty, harmlessness, power-seeking, and more, demonstrating the promise of top-down transparency research. We hope that this work catalyzes further exploration of RepE and fosters advancements in the transparency and safety of AI systems.

翻译：本文识别并阐述了表征工程（RepE）这一新兴领域，该领域借鉴认知神经科学的研究视角，旨在增强人工智能系统的透明度。表征工程将群体级表征（而非神经元或电路）置于分析核心，为我们提供了监控与操控深度神经网络（DNNs）中高层级认知现象的新方法。我们提供了RepE技术的基线方法与初步分析，表明其能为理解和控制大型语言模型提供简洁有效的解决方案。我们展示了这些方法如何解决包括诚实性、无害性、权力追求等在内的广泛安全相关问题，论证了自上而下透明度研究的潜力。希望本研究能推动RepE的深入探索，促进人工智能系统透明度与安全性的进步。