Computer Science > Computation and Language
[Submitted on 30 Oct 2026]
Title:Circuit Oracle: Automating Attribution Graph Analysis via Natural-Language Queries
View PDF HTML (experimental)Abstract:Attribution graphs, an emerging tool in mechanistic interpretability, use transcoders to decompose language model computations into sparse interpretable features connected by causal edges. However, turning a graph into a safety-relevant insight requires hours of manual analysis by experts. We introduce Circuit Oracle, an off-the-shelf LLM equipped with a multi-agent harness and task-specific authored skills, which answers natural-language questions about a target model by reading the attribution graph through tool calls. We evaluate it on three safety-relevant proxy tasks: detecting spurious features in probe circuits, eliciting hidden knowledge from taboo-finetuned models, and suppression jailbreaking via feature-level interventions. Beyond task scores, the oracle produces pruned subgraphs and context-specific feature explanations that we assess qualitatively. Results are mixed. On spurious-feature detection, Circuit Oracle reaches 78.2 $\pm$ 2.4% accuracy, outperforming single-layer feature-ranking baselines. On secret elicitation, trained self-explainers like Activation Oracles remain the stronger baseline. On refusal suppression over 50 prompts, its single committed intervention is slightly below a diff-in-means baseline, and only exceeds it when allowed the best of five attempts. Results are stable across five runs and across model choices for the agents. We also release the pruned subgraphs and context-specific feature readings that revise existing global labels, and invite future work to develop evaluations that can measure their usefulness directly.
Loading...
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
ScienceCast (What is ScienceCast?)
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.