[2604.07615] ADAG: Automatically Describing Attribution Graphs

Abstract:In language model interpretability research, \textbf{circuit tracing} aims to identify which internal features causally contributed to a particular output and how they affected each other, with the goal of explaining the computations underlying some behaviour. However, all prior circuit tracing work has relied on ad-hoc human interpretation of the role that each feature in the circuit plays, via manual inspection of data artifacts such as the dataset examples the component activates on. We introduce \textbf{ADAG}, an end-to-end pipeline for describing these attribution graphs which is fully automated. To achieve this, we introduce \textit{attribution profiles} which quantify the functional role of a feature via its input and output gradient effects. We then introduce a novel clustering algorithm for grouping features, and an LLM explainer--simulator setup which generates and scores natural-language explanations of the functional role of these feature groups. We run our system on known human-analysed circuit-tracing tasks and recover interpretable circuits, and further show ADAG can find steerable clusters which are responsible for a harmful advice jailbreak in Llama 3.1 8B Instruct.

View PDF HTML (experimental) Abstract:In language model interpretability research, \textbf{circuit tracing} aims to identify which internal features causally contributed to a particular output and how they affected each other, with the goal of explaining the computations underlying some behaviour. However, all prior circuit tracing work has relied on ad-hoc human interpretation of the role that each feature in the circuit plays, via manual inspection of data artifacts such as the dataset examples the component activates on. We introduce \textbf{ADAG}, an end-to-end pipeline for describing…

saved by

• Kushal Thaman

related reading

• Circuit Tracing: Revealing Computational Graphs in Language Modelstransformer-circuits.pub
• Towards Automated Circuit Discovery for Mechanistic Interpretabilityarxiv.org
• On the Biology of a Large Language Modeltransformer-circuits.pub
• Natural Language Autoencoders Produce Unsupervised Explanations of LLM Activationstransformer-circuits.pub
• Transformer Circuits Threadtransformer-circuits.pub
• Language Model Circuits Are Sparse in the Neuron Basis | Transluce AItransluce.org
• Circuit Tracing in Vision–Language Models:Understanding the Internal Mechanisms of Multimodal Thinkingarxiv.org
• Making LLMs Say What They Think: Measuring and Improving CoT-Interpretability Alignmentarxiv.org
• Bridging the Attention Gap: Complete Replacement Models for Complete Circuit Tracinginterp.open-moss.com
• [2604.16812] Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
• Audit Without Verification: When LLM Accountability Layers Relay Rather Than Checkarxiv.org
• [2511.08579] Training Language Models to Explain Their Own Computationsarxiv.org