Neel Nanda
Source-listed: Staff Research Scientist; lead of the mechanistic interpretability team · Google DeepMind
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- Progress measures for grokking via mechanistic interpretabilitySaved credit: First author · 2023Explore people connected to this work →
- TransformerLensSaved credit: Creator · 2022Explore people connected to this work →
- AtP*: An efficient and scalable method for localizing LLM behaviour to componentsSaved credit: Contributor · 2024Explore people connected to this work →
- Improving Dictionary Learning with Gated Sparse AutoencodersSaved credit: Contributor · 2024Explore people connected to this work →
- Open Problems in Mechanistic InterpretabilitySaved credit: Contributor · 2025Explore people connected to this work →