<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Forrest Ye — 技术与产品笔记</title><description>关于大模型、产品和真实使用的长期笔记。</description><link>https://forrestye.space/</link><language>zh-CN</language><item><title>用了半年 Meta Ray-Ban Display：它不是终局，但已经改变了我记录生活的方式</title><link>https://forrestye.space/writing/meta-ray-ban-display-six-months/</link><guid isPermaLink="true">https://forrestye.space/writing/meta-ray-ban-display-six-months/</guid><description>戴了半年 Meta Ray-Ban Display，我用它记录生活、演讲提词、听歌和看消息。它很重、国内生态也不完整，却依然是现阶段最适合我的智能眼镜。</description><pubDate>Fri, 31 Jul 2026 00:00:00 GMT</pubDate></item><item><title>DeepSeek狂飙：技术与八卦齐飞的背后真相</title><link>https://forrestye.space/writing/deepseek-v3-r1-context/</link><guid isPermaLink="true">https://forrestye.space/writing/deepseek-v3-r1-context/</guid><description>写于 DeepSeek V3 与 R1 引发集中讨论之时，尝试把模型路线、算力争议、开源意义和市场情绪放在同一张图里理解。</description><pubDate>Tue, 28 Jan 2025 00:00:00 GMT</pubDate></item><item><title>LLaMA3：开启大语言模型的新篇章</title><link>https://forrestye.space/writing/llama-3-new-chapter/</link><guid isPermaLink="true">https://forrestye.space/writing/llama-3-new-chapter/</guid><description>一篇写于 LLaMA 3 发布当天的技术笔记，关注 tokenizer、GQA、上下文长度、训练数据规模和小模型训练趋势。</description><pubDate>Sat, 20 Apr 2024 00:00:00 GMT</pubDate></item></channel></rss>