<?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>AI/ML on 3rd layer</title><link>https://3rdlayer.uk/ko/categories/ai/ml/</link><description>Recent content in AI/ML on 3rd layer</description><generator>Hugo -- 0.154.5</generator><language>ko-KR</language><lastBuildDate>Sun, 16 Feb 2025 00:00:00 +0000</lastBuildDate><atom:link href="https://3rdlayer.uk/ko/categories/ai/ml/index.xml" rel="self" type="application/rss+xml"/><item><title>Sparse Autoencoder로 GPT-2의 감정을 조종하기</title><link>https://3rdlayer.uk/ko/posts/sae-steering-gpt2/</link><pubDate>Sun, 16 Feb 2025 00:00:00 +0000</pubDate><guid>https://3rdlayer.uk/ko/posts/sae-steering-gpt2/</guid><description>OpenAI의 Pretrained SAE로 GPT-2 내부의 감정 Feature를 찾아내고, 직접 SAE를 학습하는 과정까지. Feature Patching으로 &amp;lsquo;good person&amp;rsquo;을 &amp;lsquo;shit&amp;rsquo;로 바꾸는 실험.</description></item></channel></rss>