<?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>Evals on 4oh4/dev</title>
    <link>https://4oh4.github.io/tags/evals/</link>
    <description>Recent content in Evals on 4oh4/dev</description>
    <generator>Hugo</generator>
    <language>en-gb</language>
    <lastBuildDate>Sat, 31 Oct 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://4oh4.github.io/tags/evals/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Testing Jev against three small LLMs on text classification and calibration</title>
      <link>https://4oh4.github.io/posts/2026-10-02-jev-vs-small-llms-text-classification/</link>
      <pubDate>Sat, 31 Oct 2026 00:00:00 +0000</pubDate>
      <guid>https://4oh4.github.io/posts/2026-10-02-jev-vs-small-llms-text-classification/</guid>
      <description>I compared Jev, which TypeSafe AI describes as calibrated, with three small LLMs. It was cheaper and competitive on accuracy, but no better calibrated.</description>
    </item>
  </channel>
</rss>
