Obtaining human-interpretable explanations of large, general-purpose language models is an urgent goal for AI safety. However, it is just as important that our interpretability methods are faithful to the causal dynamics underlying model behavior and able to robustly generalize to unseen inputs. Distributed Alignment Search (DAS) is a powerful gradient descent method grounded in a theory of causal abstraction that uncovered perfect alignments between interpretable symbolic algorithms and small deep learning models fine-tuned for specific tasks. In the present paper, we scale DAS significantly by replacing the remaining brute-force search steps with learned parameters -- an approach we call DAS. This enables us to efficiently search for interpretable causal structure in large language models while they follow instructions. We apply DAS to the Alpaca model (7B parameters), which, off the shelf, solves a simple numerical reasoning problem. With DAS, we discover that Alpaca does this by implementing a causal model with two interpretable boolean variables. Furthermore, we find that the alignment of neural representations with these variables is robust to changes in inputs and instructions. These findings mark a first step toward deeply understanding the inner-workings of our largest and most widely deployed language models.

翻译：获取大型通用语言模型的人类可解释性解释是人工智能安全领域的迫切目标。然而，同样重要的是，我们的可解释性方法必须忠实于模型行为背后的因果动态，并能够稳健地泛化到未见输入。分布式对齐搜索（DAS）是一种基于因果抽象理论的强大梯度下降方法，它揭示了可解释符号算法与针对特定任务微调的小型深度学习模型之间的完美对齐。在本文中，我们通过用学习到的参数替换剩余的暴力搜索步骤，显著扩展了DAS——我们将这种方法称为DAS。这使我们能够在大型语言模型遵循指令时高效地搜索其可解释的因果结构。我们将DAS应用于Alpaca模型（7B参数），该模型在开箱即用状态下能解决一个简单的数值推理问题。通过DAS，我们发现Alpaca通过实现一个包含两个可解释布尔变量的因果模型来完成此任务。此外，我们观察到神经表征与这些变量的对齐对输入和指令的变化具有鲁棒性。这些发现标志着向深入理解我们最大且最广泛部署的语言模型内部工作机制迈出了第一步。