Data-driven Circuit Discovery for Interpretability of Language Models

arXiv:2605.09129 · cs.AI · Submitted 2026-05-09 · Read on arXiv

cs.AI

Submitted: 2026-05-09

Updated: 2026-10-12

Code: https://github.com/Ziyu-Yao-NLP-Lab/data-driven-circuit-discovery

Terminology

• Data-driven Circuit Discovery — 2× in this paper
• Interpretability — 1× in this paper · explained in 15 episodes

Sources

• Mechanistic Interpretability for AI Safety -- A Review
• Summing Up the Facts: Additive Mechanisms Behind Factual Recall in LLMs
• A Primer on the Inner Workings of Transformer-based Language Models
• Finding Interpretable Prompt-Specific Circuits in Language Models
• Mixing Mechanisms: How Language Models Retrieve Bound Entities In-Context
• A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models
• Towards Best Practices of Activation Patching in Language Models: Metrics and Methods

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• Agents' Last Exam
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems