<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Representations on mu0</title><link>https://mu0.ai/tags/representations/</link><description>Recent content in Representations on mu0</description><generator>Hugo -- 0.152.2</generator><language>en-us</language><lastBuildDate>Tue, 20 Oct 2026 09:00:00 +1000</lastBuildDate><atom:link href="https://mu0.ai/tags/representations/index.xml" rel="self" type="application/rss+xml"/><item><title>What If We Could Ask a Model for a Concept?</title><link>https://mu0.ai/posts/asking-a-model-for-a-concept/</link><pubDate>Tue, 20 Oct 2026 09:00:00 +1000</pubDate><guid>https://mu0.ai/posts/asking-a-model-for-a-concept/</guid><description>A proposal: extract one vector per concept from vision, language, and audio models, align the resulting spaces using shared concept cards, and test whether the alignment generalises to concepts it never saw — before asking whether models should be moved toward a shared space at all.</description></item></channel></rss>