<?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>Series on Sana Fayyaz</title><link>https://sanafayyaz315.github.io/series/</link><description>Recent content in Series on Sana Fayyaz</description><generator>Hugo</generator><language>en-us</language><atom:link href="https://sanafayyaz315.github.io/series/index.xml" rel="self" type="application/rss+xml"/><item><title>Inside the Transformer</title><link>https://sanafayyaz315.github.io/series/inside-the-transformer/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://sanafayyaz315.github.io/series/inside-the-transformer/</guid><description>&lt;p>Transformer architecture, taken apart piece by piece — attention, positional encoding, feedforward layers, normalization, and everything else under the hood. Each post is one concept, written the way I wish someone had explained it to me: intuition first, then the mechanics, then the math.&lt;/p>
&lt;p>Posts in this series appear below as I write them.&lt;/p></description></item></channel></rss>