Mechanistic Interpretability

Ask about this topic

Overview

The field seeks to explain how a model’s internal computations produce particular outputs, rather than relying only on input-output behavior. Researchers use sparse autoencoders and transcoders to identify features, and probes, attribution graphs, activation patching, ablation, and steering to trace or test their roles. Studies examine language-model tasks such as factual recall, reasoning, and refusal, as well as vision, audio, and scientific models. A central challenge is that a signal can be easy to decode without being causally used, while circuit explanations and interventions can vary across prompts, models, and analysis choices.

Latest papers

CardsList

391 papers63 papers in the last four weeks, up 75% on the four weeks before. 0.4% of all new papers.

  1. Domain-Filtered Knowledge Graphs from Sparse Autoencoder Features

    Apr 26, 2026John Winnicki, Abeynaya Gnanasekaran, Eric DarveKG ConstructionLLM Interpretability

    Institute for Computational and Mathematical Engineering, Stanford University, Stanford, CA, USA

  2. Make Mechanistic Interpretability Auditable: A Call to Develop Guidelines via Continuous Collaborative Reviewing

    Apr 24, 2026Michael Lan, Narmeen Fatimah Oozeer, Chaithanya Bandi +4Scholarly Peer ReviewMechanistic Interpretability

    Martian · University of Delaware · MATS +2

  3. Fine-Grained Analysis of Shared Syntactic Mechanisms in Language Models

    Apr 24, 2026Ryoma Kumon, Hitomi YanakaMechanistic InterpretabilityLanguage Modeling

    1The University of Tokyo · 2RIKEN · 3Tohoku University

  4. Shared Lexical Task Representations Explain Behavioral Variability In LLMs

    Apr 23, 2026Zhuonan Yang, Jacob Xiaochen Li, Francisco Piedrahita Velez +5Prompt SensitivityLLM Prompting

    Brown University · MIT · Northeastern

  5. Mechanistic Interpretability Tool for AI Weather Models

    Apr 22, 2026Kirsten I. Tempest, Matthias Beylich, George C. CraigInterpretable MLMechanistic Interpretability

    Meteorological Institute Munich, Ludwig-Maximilians-Universität, 80333 Munich, Germany · Deutsches Zentrum für Luft- und Raumfahrt, Oberpfaffenhofen, Germany

  6. LLMs Know They're Wrong and Agree Anyway: The Shared Sycophancy-Lying Circuit

    Apr 21, 2026Manav PandeyLLM SycophancyLLM Alignment

    Georgia Institute of Technology

  7. Where Fake Citations Are Made: Tracing Field-Level Hallucination to Specific Neurons in LLMs

    Apr 20, 2026Yuefei Chen, Yihao Quan, Xiaodong Lin +1LLM Hallucination MitigationHallucination in Language Models

    Rutgers University

  8. A Mechanistic Study of Algorithmic Generalization in Diffusion Models on Modular Arithmetic

    Apr 20, 2026Joon Hyeok Kim, Yong-Hyun Park, Mattis Dalsætra Østby +1Pixel-Space Diffusion ModelsMechanistic Interpretability

    Department of Computer and Information Science, University of Pennsylvania, Philadelphia, PA 19104, USA

  9. Prune, Interpret, Evaluate: A Cross-Layer Transcoder-Native Framework for Efficient Circuit Discovery via Feature Attribution

    Apr 18, 2026Qinhao Chen, Linyang He, Nima MesgaraniLLM PruningFeature Attribution

    Columbia University

  10. Surgical Repair of Insecure Code Generation in LLMs

    Apr 17, 2026Gustavo Sandoval, Brendan Dolan-Gavitt, Siddharth GargMechanistic InterpretabilitySecure Code Generation

    Tandon School of Engineering New York University · XBOW

  11. Disentangling Mathematical Reasoning in LLMs: A Methodological Investigation of Internal Mechanisms

    Apr 17, 2026Tanja Baeumel, Josef van Genabith, Simon OstermannLLM InterpretabilityMechanistic Interpretability

    German Research Center for Artificial Intelligence (DFKI) · Saarland University · Center for European Research in Trusted AI (CERTAIN)

  12. What Is the Minimum Architecture for Prolepsis? Early Irrevocable Commitment Across Tasks in Small Transformers

    Apr 16, 2026Éric JacopinTransformer InterpretabilityAttention Head Analysis

    Cosmic AI

  13. A Mechanistic Account of Attention Sinks in GPT-2: One Circuit, Broader Implications for Mitigation

    Apr 16, 2026Yuval Ran-Milo, Hila Ofek, Shahar MendelSelf-AttentionDecoder-Only Language Models

    Tel Aviv University

  14. Seeing Through Circuits: Faithful Mechanistic Interpretability for Vision Transformers

    Apr 15, 2026Nina Żukowska, Wolfgang Stammer, Bernt Schiele +1Transformer InterpretabilityMechanistic Interpretability

    Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbrücken, Germany

  15. Weight Patching: Toward Source-Level Mechanistic Localization in LLMs

    Apr 15, 2026Chenghao Sun, Chengsheng Zhang, Guanzheng Qin +2Mechanistic InterpretabilityCausal Interventions in Language Models

    Key Laboratory of Ministry of Education for Brain Inspired Intelligent Perception and Cognition, University of Science and Technology of China, Hefei, Anhui 230026, China

  16. Distinct mechanisms underlying in-context learning in transformers

    Apr 14, 2026Cole Gibson, Wenping Cui, Gautam ReddyIn-Context LearningTransformer Training Dynamics

    Lewis-Sigler Institute for Integrative Genomics, Princeton University, Princeton, NJ 08544, USA · Department of Physics, Princeton University, Princeton, NJ 08544, USA

  17. The Linear Centroids Hypothesis: Features as Directions Learned by Local Experts

    Apr 13, 2026Thomas Walker, Ahmed Imtiaz Humayun, Randall Balestriero +1Neural Network InterpretabilityMechanistic Interpretability

    Rice University · Google Research · Brown University

  18. Sparse Autoencoder Decomposition of Clinical Sequence Model Representations: Feature Complexity, Task Specialisation, and Mortality Prediction

    Apr 13, 2026Chris Sainsbury, Feng Dong, Andreas KarwathEHR Foundation ModelsSparse Autoencoders

    School of Cardiovascular & Metabolic Health, University of Glasgow, UK · School of Medicine, University of Dundee, UK · NHS Greater Glasgow and Clyde, UK +2

  19. Do LLMs Know Tool Irrelevance? Demystifying Structural Alignment Bias in Tool Invocations

    Apr 13, 2026Yilong Liu, Xixun Lin, Pengfei Cao +3LLM Tool UseTool-Use Evaluation

    Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China · School of Cyber Security, University of Chinese Academy of Sciences, Beijing, China · Institute of Automation, Chinese Academy of Sciences, Beijing, China +1

  20. Pando: Do Interpretability Methods Work When Models Won't Explain Themselves?

    Apr 13, 2026Ziqian Zhong, Aashiq Muhamed, Mona T. Diab +2Mechanistic InterpretabilityExplainability Evaluation

    Carnegie Mellon University

  21. Tracing the Thought of a Grandmaster-level Chess-Playing Transformer

    Apr 11, 2026Rui Lin, Zhenyu Jin, Guancheng Zhou +7Transformer InterpretabilityMechanistic Interpretability

    Shanghai Innovation Institute, Shanghai, China · School of Computer Science, Fudan University, Shanghai, China · School of Mathematics and Statistics, Xi’an Jiaotong University.

  22. Functional Emotions or Situational Contexts? A Discriminating Test from the Mythos Preview System Card

    Apr 9, 2026Hiranya V. PeirisEmotion Representation in Language ModelsLanguage Model Auditing

    Institute of Astronomy & Kavli Institute for Cosmology, University of Cambridge, Madingley Road, Cambridge CB3 0HA, UK · Cavendish Laboratory, Department of Physics, University of Cambridge, JJ Thomson Avenue, Cambridge, CB3 0HE, UK

  23. Latent Anomaly Knowledge Excavation: Unveiling Sparse Sensitive Neurons in Vision-Language Models

    Apr 9, 2026Shaotian Li, Shangze Li, Chuancheng Shi +5Multimodal Anomaly DetectionZero-Shot Anomaly Detection

    Macquarie University · Nanjing University of Science and Technology · The University of Sydney +1

  24. ADAG: Automatically Describing Attribution Graphs

    Apr 8, 2026Aryaman Arora, Zhengxuan Wu, Jacob Steinhardt +1Mechanistic InterpretabilityCircuits in Language Models

    Stanford University · Transluce

  25. ConceptTracer: Interactive Analysis of Concept Saliency and Selectivity in Neural Representations

    Apr 8, 2026Ricardo Knauer, Andre Beinrucker, Erik RodnerNeural Network InterpretabilityNeuron-Level Interpretability

    KI-Werkstatt, University of Applied Sciences Berlin, Berlin, Germany

  26. Improving Robustness In Sparse Autoencoders via Masked Regularization

    Apr 7, 2026Vivek Narayanaswamy, Kowshik Thopalli, Bhavya Kailkhura +1Sparse AutoencodersMechanistic Interpretability

    Lawrence Livermore National Laboratory, CA, USA

  27. MechELK: A Mechanistic Interpretability Framework for Eliciting Latent Knowledge in Large Language Models

    Apr 7, 2026Ji-jun Park, Soo-joon Choi, Jiwon Jeong +2LLM InterpretabilityLanguage Model Probing

    Dongguk University

  28. Disentangling MLP Neuron Weights in Vocabulary Space

    Apr 7, 2026Asaf Avrahamy, Yoav Gur-Arieh, Mor GevaLLM InterpretabilityTransformer Interpretability

    Blavatnik School of Computer Science and AI, Tel Aviv University

  29. Shifting the Gradient: Understanding How Defensive Training Methods Protect Language Model Integrity

    Apr 3, 2026Satchel Grant, Victor Gillioz, Jake Ward +1Adversarial Robustness of Language ModelsMechanistic Interpretability

    Stanford University & MATS · MATS · Goodfire

  30. CRaFT: Circuit-Guided Refusal Feature Selection via Cross-Layer Transcoders

    Apr 2, 2026Su-Hyeon Kim, Hyundong Jin, Yejin Lee +1Refusal Behavior in Language ModelsMechanistic Interpretability

    Yonsei University, Seoul, Republic of Korea