[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-71411":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":17,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":10,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":23,"hasPages":21,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":16,"starSnapshotCount":16,"syncStatus":28,"lastSyncTime":29,"discoverSource":30},71411,"llm-viz","bbycroft\u002Fllm-viz","bbycroft","3D Visualization of an GPT-style LLM","https:\u002F\u002Fbbycroft.net",null,"TypeScript",5408,642,42,9,0,5,24,15,39.42,false,"main",true,[],"2026-06-12 02:02:52","\n# Brendan Bycroft's Home Page & Projects\n\nThis repository contains my (Brendan's) homepage, as well as a number of non-trivial projects.\n\nThey are kept in a single repository for ease of deployment, as well as sharing a bunch of js utils\nwhich are otherwise a pain to share around.\n\n## Projects\n\nThe main projects are:\n* LLM Visualization: 3D interactive model of a GPT-style LLM network running inference.\n* [WIP] CPU Simulation: A 2D digital schematic editor with full a execution model, showcasing a simple\n  RISC-V based CPU\n\n### LLM Visualization\n\nThis project displays a 3D model of a working implementation of a GPT-style network. That\nis, the network topology that's used in OpenAI's GPT-2, GPT-3, (and maybe GPT-4).\n\nThe first network displayed with working weights is a tiny such network, which sorts a small list\nof the letters A, B, and C. This is the demo example model from Andrej Karpathy's\n[minGPT](https:\u002F\u002Fgithub.com\u002Fkarpathy\u002FminGPT) implementation.\n\nThe renderer also supports visualizing arbitrary sized networks, and works with the smaller gpt2\nsize, although the weights aren't downloaded (it's 100's of MBs).\n\n### CPU Simulation (WIP; not exposed yet!)\n\nThis project runs 2D schematic digital circuits, with a fully fledged editor. The intent is to\nadd a number of walkthroughs, showing things such as:\n  * how a simple RISC-V CPU is constructed\n  * the constituent parts down to gate level: instruction decode, ALU, add, etc\n  * higher level CPU ideas, like various levels of pipelining, caching, etc\n\n## Running Locally\n\n1. Install dependencies: `yarn`\n1. Start the dev server: `yarn dev`\n","llm-viz项目提供了一个GPT风格大语言模型的3D可视化交互界面。其核心功能包括展示一个工作中的小型GPT网络模型如何处理简单的排序任务，同时支持用户自定义不同规模的网络进行可视化探索，尽管对于较大的模型（如gpt2），由于文件体积过大，并不会自动下载权重文件。该项目使用TypeScript编写，适合于对深度学习尤其是Transformer架构有兴趣的研究者、开发者或教育工作者，在教学、研究或是自我学习时更直观地理解神经网络的工作机制。",2,"2026-06-11 03:37:37","high_star"]