<?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>微调 on 技术博客 - 有条工具</title><link>/blog/tags/%E5%BE%AE%E8%B0%83/</link><description>Recent content in 微调 on 技术博客 - 有条工具</description><generator>Hugo -- 0.147.7</generator><language>zh-cn</language><copyright>2024-2025 有条工具技术博客</copyright><lastBuildDate>Sat, 03 Oct 2026 10:05:00 +0800</lastBuildDate><atom:link href="/blog/tags/%E5%BE%AE%E8%B0%83/feed.xml" rel="self" type="application/rss+xml"/><item><title>LoRA 微调入门：单卡调出自己的模型</title><link>/blog/articles/lora-qlora-finetune-guide/</link><pubDate>Sat, 03 Oct 2026 10:05:00 +0800</pubDate><guid>/blog/articles/lora-qlora-finetune-guide/</guid><description>从数据准备到训练与合并，走通 LoRA/QLoRA 微调全流程。核心原则是先算显存预算再选方案，避免一上来就撞 OOM。</description></item></channel></rss>