<?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>FP16 on Warpspace Blog</title><link>http://blog.caveduck.io/tags/fp16/</link><description>Recent content in FP16 on Warpspace Blog</description><generator>Hugo -- 0.154.5</generator><language>en-US</language><lastBuildDate>Mon, 13 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="http://blog.caveduck.io/tags/fp16/index.xml" rel="self" type="application/rss+xml"/><item><title>Efficient ML #1 — Deep Learning Data Types</title><link>http://blog.caveduck.io/posts/deep-learning-data-types/</link><pubDate>Mon, 13 Apr 2026 00:00:00 +0000</pubDate><guid>http://blog.caveduck.io/posts/deep-learning-data-types/</guid><description>INT8, FP16, BF16, FP8 (E4M3/E5M2), FP4 — what do all these data types flooding deep learning model cards actually mean, and how do bits get interpreted as numbers? We take them apart one by one, with a widget where clicking bits updates the formula and value in real time.</description></item></channel></rss>