Stefan Heimersheim

1. Open Problems in Mechanistic Interpretability
   2025/01/27 by Lee Sharkey, Bilal Chughtai, Sharkey, Lee +58 · 7 voices · 38 citations
   Computer Science · #Natural Language Processing Techniques #Statistical and Computational Modeling #cs.LG
2. Towards Automated Circuit Discovery for Mechanistic Interpretability
   2023/04/28 by Arthur Conmy, Augustine N. Mavor-Parker, Conmy, Arthur +7 · 81 citations
   Computer Science · Engineering · #Adversarial Robustness in Machine Learning #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Fault Detection and Control Systems #Machine Learning (cs.LG)
3. How to use and interpret activation patching
   2024/04/23 by Stefan Heimersheim, Neel Nanda, Heimersheim, Stefan +1 · 36 citations
   Computer Science · Business, Management and Accounting · #Usability and User Interface Design #Business Process Modeling and Analysis #Intelligent Tutoring Systems and Adaptive Learning
4. Detecting Strategic Deception Using Linear Probes
   2025/02/05 by Nicholas Goldowsky-Dill, Bilal Chughtai, Goldowsky-Dill, Nicholas +5 · 22 citations
   Computer Science · Psychology · #Advanced Malware Detection Techniques #Deception detection and forensic psychology #FOS: Computer and information sciences #Machine Learning (cs.LG) #Network Security and Intrusion Detection
5. Interpretability in Parameter Space: Minimizing Mechanistic Description Length with Attribution-based Parameter Decomposition
   2025/01/24 by Dan Braun, Lucius Bushnaq, Braun, Dan +7 · 9 citations
   Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Natural Language Processing Techniques
6. You can remove GPT2's LayerNorm by fine-tuning
   2024/09/06 by Stefan Heimersheim, Heimersheim, Stefan · 3 citations
   Computer Science · #Advanced Data Storage Technologies #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Parallel Computing and Optimization Techniques
7. Rapid and Late Cosmic Reionization Driven by Massive Galaxies: a Joint Analysis of Constraints from 21-cm, Lyman Line & CMB Data Sets
   2025/04/13 by Peter Sims, Sims, Peter H., H. T. J. Bevins +15 · 3 citations
   Physics and Astronomy · #Astronomy and Astrophysical Research #Cosmology and Nongalactic Astrophysics (astro-ph.CO) #FOS: Physical sciences #Instrumentation and Methods for Astrophysics (astro-ph.IM) #Radio Astronomy Observations and Technology
8. Benchmarking Deception Probes via Black-to-White Performance Boosts
   2025/07/16 by Avi Parrack, Parrack, Avi, Carlo Leonardo Attubato +3 · 5 citations
   Computer Science · #Advanced Malware Detection Techniques #Adversarial Robustness in Machine Learning #Network Security and Intrusion Detection
9. The Local Interaction Basis: Identifying Computationally-Relevant and Sparsely Interacting Features in Neural Networks
   2024/05/17 by Lucius Bushnaq, Stefan Heimersheim, Bushnaq, Lucius +17 · 1 citation
   Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Neural Networks and Applications
10. Using Degeneracy in the Loss Landscape for Mechanistic Interpretability
    2024/05/17 by Lucius Bushnaq, Jake Mendel, Bushnaq, Lucius +13 · 1 citation
    Computer Science · #FOS: Computer and information sciences #Machine Learning (cs.LG) #Natural Language Processing Techniques #Statistical and Computational Modeling