<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Gpt on Programmer.ie: Modern AI programming</title>
    <link>http://programmer.ie/tags/gpt/</link>
    <description>Recent content in Gpt on Programmer.ie: Modern AI programming</description>
    <generator>Hugo</generator>
    <language>en-us</language>
    <lastBuildDate>Sat, 08 Aug 2026 14:03:00 +0100</lastBuildDate>
    <atom:link href="http://programmer.ie/tags/gpt/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>PyTorch Zero to Hero 10: Build a Small GPT-Style Language Model From Scratch</title>
      <link>http://programmer.ie/post/pytorch-zero-to-hero-10/</link>
      <pubDate>Sat, 08 Aug 2026 14:03:00 +0100</pubDate>
      <guid>http://programmer.ie/post/pytorch-zero-to-hero-10/</guid>
      <description>&lt;h1 id=&#34;build-a-small-gpt-style-language-model-from-scratch-in-pytorch&#34;&gt;Build a Small GPT-Style Language Model From Scratch in PyTorch&lt;/h1&gt;&#xA;&lt;p&gt;This is the final post in the &lt;strong&gt;PyTorch: Zero to Hero&lt;/strong&gt; series.&lt;/p&gt;&#xA;&lt;p&gt;We have spent the previous posts learning the machinery underneath PyTorch:&lt;/p&gt;&#xA;&lt;ul&gt;&#xA;&lt;li&gt;tensors and shapes;&lt;/li&gt;&#xA;&lt;li&gt;autograd;&lt;/li&gt;&#xA;&lt;li&gt;manual neural networks;&lt;/li&gt;&#xA;&lt;li&gt;&lt;code&gt;nn.Module&lt;/code&gt; and parameter registration;&lt;/li&gt;&#xA;&lt;li&gt;DataLoader performance;&lt;/li&gt;&#xA;&lt;li&gt;convolutional networks;&lt;/li&gt;&#xA;&lt;li&gt;attention and masks;&lt;/li&gt;&#xA;&lt;li&gt;training failures;&lt;/li&gt;&#xA;&lt;li&gt;CUDA performance and &lt;code&gt;torch.compile&lt;/code&gt;.&lt;/li&gt;&#xA;&lt;/ul&gt;&#xA;&lt;p&gt;Now we put it together.&lt;/p&gt;&#xA;&lt;p&gt;The goal is not to download a pretrained model.&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
