← Back to PeopleResearcherChristopher PottsProfessor at Bigspin AILinkedIn@ChrisGPottsCo-authored paper on Matryoshka attribution method for language model interpretability.1Mentions1Articles1OutletsCompanies CoveredAnthropic1ArticlesMost recent firstMentioned InReferenced in coverageNew attribution method tops interpretability benchmark by 2.9x, pinpoints Llama refusals to 1% of weightsarXivpositiveSep 24, 2026“Co-authored paper on Matryoshka attribution method for language model interpretability.”