Siyuan He

AI Engineer. AI Safety & Interpretability. Learn, teach and share.

代表作品 / 项目

llm-from-scratch⭐ 182
transparent-grpo⭐ 91A transparent, single-file implementation for understanding GRPO (K1 in Rewards), free from the abstractions of large libraries.
OmniTutor⭐ 38
bpe-optimized-from-scratch⭐ 9
OmniTutor_2⭐ 9

条目信息来自其公开主页与公开发布内容。按本站规范，页面不展示任何联系方式。需要更正或删除？通过收录与更正通道提交，24 小时内处理。