Looking for AI/ML posts?Browse the AI/ML section.
July 10, 2026
Reasoning models since the dawn of o1 and R1 have a tendency to overthink. Despite a lot of work on early-exit methods and steering, open-weight and smaller reasoning models still produce long chains of thought before they answer. I worked on discovering how much of that thinking is required before the model already knows the answer and what makes the model stop thinking.
June 27, 2026
Linear probes achieve near-perfect AUC at detecting reward hacks but add almost nothing over simple output checks. An experiment with GRPO on Qwen2.5-0.5B shows the limits of mechanistic interpretability when hacks are easily verifiable from the output.
June 22, 2026
My notes and understanding based on Anthropic's 'A Mathematical Framework for Transformer Circuits' paper.
June 14, 2026
A progress report on a side project: I built a small pipeline to read the hidden states of code models, then probed four sizes of Qwen2.5-Coder. Scale makes them settle the grammar earlier in the network and barely changes how well they track what a program does.
November 27, 2025
My first blog post where I share what this blog will be about.