phenomenon · first seen 15 Oct 2026
polysemanticity
The tendency of individual neurons to respond to multiple unrelated concepts, complicating interpretability.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Explaining the Roles of Experts in Mixture-of-Experts LLMs for PersonalizationPapers · 15 Oct 2026 · Sekh Mainul Islam, Pepa Atanasova, Isabelle Augenstein
- Comparing Sparsified MLPs and Transcoders: Many Transcoder Features Are Neurons in DisguisePapers · 17 Oct 2026 · Stepan Shabalin, Mateusz Piotrowski
- Understanding AI from the Inside: The State of Mechanistic Interpretability for AI SafetyPapers · 5 Nov 2026 · Fazl Barez, Tung-Yu Wu, Hunar Batra, Gabriele Dominici, Florent Draye, Yanai Ela…
Latest sources
- Understanding AI from the Inside: The State of Mechanistic Interpretability for AI SafetyPapers · 5 Nov 2026 · Fazl Barez, Tung-Yu Wu, Hunar Batra, Gabriele Dominici, Florent Draye, Yanai Ela…
- Comparing Sparsified MLPs and Transcoders: Many Transcoder Features Are Neurons in DisguisePapers · 17 Oct 2026 · Stepan Shabalin, Mateusz Piotrowski
- Explaining the Roles of Experts in Mixture-of-Experts LLMs for PersonalizationPapers · 15 Oct 2026 · Sekh Mainul Islam, Pepa Atanasova, Isabelle Augenstein