I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders
Paper
•
2503.18878
•
Published
•
119
Interpretability
-
-
Truth Neurons
Paper • 2505.12182 • Published • 8 -
Resa: Transparent Reasoning Models via SAEs
Paper • 2506.09967 • Published • 21 -
Why Can't Transformers Learn Multiplication? Reverse-Engineering Reveals Long-Range Dependency Pitfalls
Paper • 2510.00184 • Published • 16 -
Mechanistic Interpretability of Large-Scale Counting in LLMs through a System-2 Strategy
Paper • 2601.02989 • Published • 4