Mechanistic interpretability researchers applying causality theory to LLMs

Related Stories

• Memo: Tang Jie, founder of Z.ai, the Chinese lab behind the GLM models, argues frontier AI capabilities should stay "as open and widely accessible as possible"
• Sources: Cursor is building a general-purpose AI agent codenamed Sand, aimed at non-developers, that handles emails, texts, and documents to rival Claude Cowork
• OpenAI, Meta, and SpaceXAI may be able to put pressure on Anthropic by emphasizing cost efficiency, as business customers increasingly scrutinize AI spending
• Lorde says Ray-Ban Meta AI glasses are ‘not sexy’
• I love LLMs, I hate hype