Blog

Why. This blog is for sharing my ongoing experiments and observations — mostly around mechanistic interpretability and alignment in language models. I am starting it to make my work more accessible and tangible, beyond code or internal notes (which are often in rough shape). I also hope that writing here helps me stay accountable, focus on clarity, and continually improve how I communicate research.

Posts

Review System

15.10.25 •

Thoughts on feedback and a review system for future blogs.

“A meta-analysis on the value of critical feedback culminates in the proposal of an ‘Egg Rating’ system driven by anti-sycophantic LLM prompts.” ~Gemini