{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreieu3iaozzrlcgkzuzs4afxylqn7umnrovmjub2nqna6nmne6uodma",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mpfpsep6x4h2"
  },
  "path": "/t/a-comprehensive-bilingual-guide-to-transformers-from-foundations-to-kv-cache-compression-attention-dynamics/177222#post_1",
  "publishedAt": "2026-06-29T03:54:17.000Z",
  "site": "https://discuss.huggingface.co",
  "tags": [
    "https://karlesmarin.github.io/transformers-guide/en/index.html",
    "Cómo Atienden los Transformers"
  ],
  "textContent": "Hi everyone,\n\nI’d like to share an open-source resource I’ve been working on: a comprehensive, bilingual (English & Spanish) guide on Transformer architectures.\n\nMy goal was to create a bridge between the mathematical foundations of attention mechanisms and their practical implementation. The guide goes beyond the basics and dives deep into low-level mechanics, including:\n\n  * **Attention Dynamics:** From scratch implementations to understanding attention collapse.\n\n  * **Context & Memory:** Exploring KV-cache compression and long-context challenges.\n\n  * **Advanced Concepts:** Grokking, optimization, and structural analysis.\n\n\n\n\nThe theoretical explanations are backed by reproducible code and interactive elements (like the TAF Agent framework I’ve been developing for browser-based LLM testing).\n\nYou can read it here:  English: https://karlesmarin.github.io/transformers-guide/en/index.html _(Nota: asegúrate de que esta URL existe o cámbiala por la correcta)_ Spanish: Cómo Atienden los Transformers\n\nI’d love to hear feedback from the community, especially regarding the visualization of attention states and optimization techniques. Contributions or suggestions are more than welcome!\n\nCheers, Carles",
  "title": "A comprehensive, bilingual guide to Transformers: From foundations to KV-cache compression & attention dynamics"
}