[ Mechanistic-Interpretability ]

Discover insights, tutorials, and thoughts on technology, homelab, and development.

Apr 2, 2026 10 min read

Topology of Thought

How curiosity, physics, and looking inside neural networks revealed something unexpected about how computation organizes itself

ai-research topology persistent-homology

Mar 23, 2026 15 min read

How we solved all 3 backdoored DeepSeek V3 models using SVD weight analysis, persistent homology, multi-model deliberation, and 5,000+ indexed probes

ai-research machine-learning security