<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Mechanistic Interpretability | Georg Lange</title><link>https://georglange.com/tags/mechanistic-interpretability/</link><atom:link href="https://georglange.com/tags/mechanistic-interpretability/index.xml" rel="self" type="application/rss+xml"/><description>Mechanistic Interpretability</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Wed, 09 Apr 2025 00:00:00 +0000</lastBuildDate><image><url>https://georglange.com/media/icon_hu12340157806326524241.png</url><title>Mechanistic Interpretability</title><link>https://georglange.com/tags/mechanistic-interpretability/</link></image><item><title>Towards principled evaluations of sparse autoencoders for interpretability and control</title><link>https://georglange.com/publication/sae-evals/</link><pubDate>Wed, 09 Apr 2025 00:00:00 +0000</pubDate><guid>https://georglange.com/publication/sae-evals/</guid><description/></item><item><title>Is This the Subspace You Are Looking For? An Interpretability Illusion for Subspace Activation Patching</title><link>https://georglange.com/publication/conference-paper/</link><pubDate>Thu, 09 May 2024 00:00:00 +0000</pubDate><guid>https://georglange.com/publication/conference-paper/</guid><description/></item></channel></rss>