<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Tyrone White</title><description>Tyrone White: personal site, projects, and a blog on machine learning and interpretability.</description><link>https://tyronewhite.dev/</link><item><title>Summoned by the Question</title><link>https://tyronewhite.dev/blog/summoned-by-the-question/</link><guid isPermaLink="true">https://tyronewhite.dev/blog/summoned-by-the-question/</guid><description>What mechanistic interpretability says about whether an LLM commits to a secret in 20 Questions.</description><pubDate>Sat, 30 May 2026 00:00:00 GMT</pubDate><category>interpretability</category><category>LLMs</category><category>mech-interp</category></item></channel></rss>