<?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>LLM推理 on Ringi's Log</title><link>https://lilinji.github.io/tags/llm%E6%8E%A8%E7%90%86/</link><description>Recent content in LLM推理 on Ringi's Log</description><generator>Hugo -- 0.152.2</generator><language>zh-cn</language><lastBuildDate>Fri, 21 Aug 2026 00:00:00 +0800</lastBuildDate><atom:link href="https://lilinji.github.io/tags/llm%E6%8E%A8%E7%90%86/index.xml" rel="self" type="application/rss+xml"/><item><title>AI Infra 大话西游实战之Transformer小白学习笔记（一）</title><link>https://lilinji.github.io/2026/08/ai-infra-%E5%A4%A7%E8%AF%9D%E8%A5%BF%E6%B8%B8%E5%AE%9E%E6%88%98%E4%B9%8Btransformer%E5%B0%8F%E7%99%BD%E5%AD%A6%E4%B9%A0%E7%AC%94%E8%AE%B0%E4%B8%80/</link><pubDate>Fri, 21 Aug 2026 00:00:00 +0800</pubDate><guid>https://lilinji.github.io/2026/08/ai-infra-%E5%A4%A7%E8%AF%9D%E8%A5%BF%E6%B8%B8%E5%AE%9E%E6%88%98%E4%B9%8Btransformer%E5%B0%8F%E7%99%BD%E5%AD%A6%E4%B9%A0%E7%AC%94%E8%AE%B0%E4%B8%80/</guid><description>以第一性原理视角拆解 Transformer：从 Self-Attention、QKV、FFN、RoPE、Pre-Norm 到 LLaMA-2 参数与显存手算，贯通 CUDA 算子、分布式训练与 vLLM 推理的 AI Infra 实战。</description></item></channel></rss>