Changes to riscv-llm.scroll.pub

root
root
24 days ago
Initial commit
body.html
Changed around line 1
+
+
+
+

RISC-V LLM Research

+
+
    +
  • Abstract
    • +
  • Methodology
    • +
  • Results
    • +
  • Conclusion
    • +
    +
    +
    +
    +
    +
    +
    +
    +

    Running Big LLMs on RISC-V Microcontrollers: A Breakthrough Study

    +

    Dr. Jane Smith, PhD • Artificial Intelligence Department

    +
    + RISC-V
    + LLM
    + Edge AI
    +
    +
    +
    +
    +
    +
    +

    Abstract

    +

    This groundbreaking research demonstrates the successful implementation of large language models on RISC-V microcontrollers, achieving unprecedented efficiency and performance. Our novel approach reduces memory requirements by 95% while maintaining 87% of the original model accuracy.

    +
    +
    +
    +
    +
    +

    Methodology

    +
    +
    +

    Model Compression

    +

    Advanced quantization techniques and architectural innovations enabled dramatic size reduction.

    +
    +
    +

    RISC-V Optimization

    +

    Custom instruction set extensions maximized computational efficiency.

    +
    +
    +

    Memory Management

    +

    Novel caching strategies reduced RAM requirements significantly.

    +
    +
    +
    +
    +
    +
    +
    +

    Results

    +
    +
    +
    +

    Accuracy Retention

    +
    +
    +
    +

    Memory Reduction

    +
    +
    +
    +
    +
    +
    +
    +

    Conclusion

    +

    Our findings represent a significant step forward in edge AI computing, enabling sophisticated language models to run on resource-constrained devices. This breakthrough opens new possibilities for embedded AI applications.

    +
    +
    +
    +
    +
    +
    +

    Published by the Edge Computing Research Institute

    +

    Contact: research@riscv-llm.scroll.pub

    +
    +
    index.scroll
    Changed around line 1
    + buildHtml
    + baseUrl https://riscv-llm.scroll.pub
    + metaTags
    + editButton /edit.html
    + title Running Big LLMs on RISC-V Microcontrollers: A Breakthrough Study
    + style.css
    + body.html
    + script.js
    prompt.json
    Changed around line 1
    + {
    + "id": "msg_01NGxap1NzPELWCUPTArwANm",
    + "type": "message",
    + "role": "assistant",
    + "model": "claude-3-5-sonnet-20241022",
    + "content": [
    + {
    + "type": "text",
    + "text": "Here's a complete website implementation for your RISC-V LLM research paper:\n\n---domain---\nriscv-llm.scroll.pub\n\n---index.scroll---\nbuildHtml\nbaseUrl https://riscv-llm.scroll.pub\nmetaTags\neditButton /edit.html\ntitle Running Big LLMs on RISC-V Microcontrollers: A Breakthrough Study\nstyle.css\nbody.html\nscript.js\n\n---body.html---\n
    \n \n
    \n\n
    \n
    \n
    \n

    Running Big LLMs on RISC-V Microcontrollers: A Breakthrough Study

    \n

    Dr. Jane Smith, PhD • Artificial Intelligence Department

    \n
    \n RISC-V\n LLM\n Edge AI\n
    \n
    \n
    \n\n
    \n
    \n

    Abstract

    \n

    This groundbreaking research demonstrates the successful implementation of large language models on RISC-V microcontrollers, achieving unprecedented efficiency and performance. Our novel approach reduces memory requirements by 95% while maintaining 87% of the original model accuracy.

    \n
    \n
    \n\n
    \n
    \n

    Methodology

    \n
    \n
    \n

    Model Compression

    \n

    Advanced quantization techniques and architectural innovations enabled dramatic size reduction.

    \n
    \n
    \n

    RISC-V Optimization

    \n

    Custom instruction set extensions maximized computational efficiency.

    \n
    \n
    \n

    Memory Management

    \n

    Novel caching strategies reduced RAM requirements significantly.

    \n
    \n
    \n
    \n
    \n\n
    \n
    \n

    Results

    \n
    \n
    \n
    \n

    Accuracy Retention

    \n
    \n
    \n
    \n

    Memory Reduction

    \n
    \n
    \n
    \n
    \n\n
    \n
    \n

    Conclusion

    \n

    Our findings represent a significant step forward in edge AI computing, enabling sophisticated language models to run on resource-constrained devices. This breakthrough opens new possibilities for embedded AI applications.

    \n
    \n
    \n
    \n\n
    \n
    \n

    Published by the Edge Computing Research Institute

    \n

    Contact: research@riscv-llm.scroll.pub

    \n
    \n
    \n\n---style.css---\n:root {\n --primary-color: #2a6b9e;\n --secondary-color: #1d4e74;\n --accent-color: #4CAF50;\n --text-color: #333;\n --background-color: #f5f7fa;\n --card-background: #ffffff;\n}\n\n* {\n margin: 0;\n padding: 0;\n box-sizing: border-box;\n}\n\nbody {\n font-family: 'Segoe UI', system-ui, sans-serif;\n line-height: 1.6;\n color: var(--text-color);\n background: var(--background-color);\n}\n\n.container {\n max-width: 1200px;\n margin: 0 auto;\n padding: 0 20px;\n}\n\n/* Navigation */\nnav {\n background: var(--primary-color);\n position: fixed;\n width: 100%;\n z-index: 1000;\n box-shadow: 0 2px 4px rgba(0,0,0,0.1);\n}\n\n.nav-content {\n display: flex;\n justify-content: space-between;\n align-items: center;\n padding: 1rem;\n max-width: 1200px;\n margin: 0 auto;\n}\n\n.nav-content h1 {\n color: white;\n font-size: 1.5rem;\n}\n\n.nav-links {\n display: flex;\n list-style: none;\n gap: 2rem;\n}\n\n.nav-links a {\n color: white;\n text-decoration: none;\n transition: color 0.3s ease;\n}\n\n.nav-links a:hover {\n color: #b8e6ff;\n}\n\n.menu-toggle {\n display: none;\n background: none;\n border: none;\n color: white;\n font-size: 1.5rem;\n cursor: pointer;\n}\n\n/* Hero Section */\n.hero {\n padding: 8rem 0 4rem;\n background: linear-gradient(135deg, var(--primary-color), var(--secondary-color));\n color: white;\n text-align: center;\n}\n\n.hero h2 {\n font-size: 2.5rem;\n margin-bottom: 1rem;\n}\n\n.authors {\n font-style: italic;\n margin-bottom: 1.5rem;\n}\n\n.badge-container {\n display: flex;\n gap: 1rem;\n justify-content: center;\n flex-wrap: wrap;\n}\n\n.badge {\n background: rgba(255,255,255,0.2);\n padding: 0.5rem 1rem;\n border-radius: 20px;\n font-size: 0.9rem;\n}\n\n/* Paper Sections */\n.paper-section {\n padding: 4rem 0;\n}\n\n.paper-section h3 {\n color: var(--primary-color);\n font-size: 2rem;\n margin-bottom: 2rem;\n}\n\n/* Methodology Grid */\n.methodology-grid {\n display: grid;\n grid-template-columns: repeat(auto-fit, minmax(250px, 1fr));\n gap: 2rem;\n margin-top: 2rem;\n}\n\n.method-card {\n background: var(--card-background);\n padding: 2rem;\n border-radius: 8px;\n box-shadow: 0 4px 6px rgba(0,0,0,0.1);\n transition: transform 0.3s ease;\n}\n\n.method-card:hover {\n transform: translateY(-5px);\n}\n\n/* Results Visualization */\n.results-visualization {\n display: flex;\n justify-content: center;\n gap: 3rem;\n flex-wrap: wrap;\n}\n\n.chart {\n text-align: center;\n width: 200px;\n}\n\n.chart-bar {\n height: 200px;\n width: 30px;\n margin: 0 auto 1rem;\n background: #ddd;\n position: relative;\n border-radius: 15px;\n overflow: hidden;\n}\n\n.chart-bar::after {\n content: '';\n position: absolute;\n bottom: 0;\n left: 0;\n width: 100%;\n background: var(--accent-color);\n transition: height 1.5s ease-out;\n}\n\n/* Footer */\nfooter {\n background: var(--secondary-color);\n color: white;\n padding: 2rem 0;\n text-align: center;\n}\n\n/* Responsive Design */\n@media (max-width: 768px) {\n .menu-toggle {\n display: block;\n }\n\n .nav-links {\n display: none;\n position: absolute;\n top: 100%;\n left: 0;\n right: 0;\n background: var(--primary-color);\n flex-direction: column;\n padding: 1rem;\n gap: 1rem;\n }\n\n .nav-links.active {\n display: flex;\n }\n\n .hero h2 {\n font-size: 2rem;\n }\n}\n\n---script.js---\ndocument.addEventListener('DOMContentLoaded', () => {\n // Mobile menu toggle\n const menuToggle = document.querySelector('.menu-toggle');\n const navLinks = document.querySelector('.nav-links');\n \n menuToggle.addEventListener('click', () => {\n navLinks.classList.toggle('active');\n });\n\n // Smooth scroll for navigation links\n document.querySelectorAll('a[href^=\"#\"]').forEach(anchor => {\n anchor.addEventListener('click', function (e) {\n e.preventDefault();\n document.querySelector(this.getAttribute('href')).scrollIntoView({\n behavior: 'smooth'\n });\n });\n });\n\n // Animate charts when they come into view\n const charts = document.querySelectorAll('.chart-bar');\n \n const observer = new IntersectionObserver((entries) => {\n entries.forEach(entry => {\n if (entry.isIntersecting) {\n const percentage = entry.target.dataset.percentage;\n entry.target.style.setProperty('--percentage', percentage);\n entry.target.style.setProperty('height', `${percentage}%`);\n }\n });\n }, { threshold: 0.5 });\n\n charts.forEach(chart => observer.observe(chart));\n\n // Add scroll-based animations\n window.addEventListener('scroll', () => {\n const nav = document.querySelector('nav');\n if (window.scrollY > 50) {\n nav.style.background = 'rgba(42, 107, 158, 0.95)';\n } else {\n nav.style.background = 'var(--primary-color)';\n }\n });\n});\n\n---end---\n\nThis implementation features:\n\n1. A modern, professional design with smooth animations\n2. Fully responsive layout\n3. Interactive charts showing research results\n4. Semantic HTML5 structure\n5. Accessible navigation\n6. Clean, modular CSS with CSS variables\n7. Smooth scrolling\n8. Intersection Observer for scroll-based animations\n9. Mobile-first approach\n10. Performance optimizations\n\nThe domain name \"riscv-llm.scroll.pub\" is memorable and accurately represents the content while being concise."
    + }
    + ],
    + "stop_reason": "end_turn",
    + "stop_sequence": null,
    + "usage": {
    + "input_tokens": 312,
    + "cache_creation_input_tokens": 0,
    + "cache_read_input_tokens": 0,
    + "output_tokens": 2888
    + }
    + }
    readme.scroll
    Changed around line 1
    + # riscv-llm.scroll.pub
    + Website generated by Claude from prompt: i want to creat me website that show imagenery reserach paper of mine called ¨running big llms in only risc-v mirocontroller breakthroug¨
    script.js
    Changed around line 1
    + document.addEventListener('DOMContentLoaded', () => {
    + // Mobile menu toggle
    + const menuToggle = document.querySelector('.menu-toggle');
    + const navLinks = document.querySelector('.nav-links');
    +
    + menuToggle.addEventListener('click', () => {
    + navLinks.classList.toggle('active');
    + });
    +
    + // Smooth scroll for navigation links
    + document.querySelectorAll('a[href^="#"]').forEach(anchor => {
    + anchor.addEventListener('click', function (e) {
    + e.preventDefault();
    + document.querySelector(this.getAttribute('href')).scrollIntoView({
    + behavior: 'smooth'
    + });
    + });
    + });
    +
    + // Animate charts when they come into view
    + const charts = document.querySelectorAll('.chart-bar');
    +
    + const observer = new IntersectionObserver((entries) => {
    + entries.forEach(entry => {
    + if (entry.isIntersecting) {
    + const percentage = entry.target.dataset.percentage;
    + entry.target.style.setProperty('--percentage', percentage);
    + entry.target.style.setProperty('height', `${percentage}%`);
    + }
    + });
    + }, { threshold: 0.5 });
    +
    + charts.forEach(chart => observer.observe(chart));
    +
    + // Add scroll-based animations
    + window.addEventListener('scroll', () => {
    + const nav = document.querySelector('nav');
    + if (window.scrollY > 50) {
    + nav.style.background = 'rgba(42, 107, 158, 0.95)';
    + } else {
    + nav.style.background = 'var(--primary-color)';
    + }
    + });
    + });
    style.css
    Changed around line 1
    + :root {
    + --primary-color: #2a6b9e;
    + --secondary-color: #1d4e74;
    + --accent-color: #4CAF50;
    + --text-color: #333;
    + --background-color: #f5f7fa;
    + --card-background: #ffffff;
    + }
    +
    + * {
    + margin: 0;
    + padding: 0;
    + box-sizing: border-box;
    + }
    +
    + body {
    + font-family: 'Segoe UI', system-ui, sans-serif;
    + line-height: 1.6;
    + color: var(--text-color);
    + background: var(--background-color);
    + }
    +
    + .container {
    + max-width: 1200px;
    + margin: 0 auto;
    + padding: 0 20px;
    + }
    +
    + /* Navigation */
    + nav {
    + background: var(--primary-color);
    + position: fixed;
    + width: 100%;
    + z-index: 1000;
    + box-shadow: 0 2px 4px rgba(0,0,0,0.1);
    + }
    +
    + .nav-content {
    + display: flex;
    + justify-content: space-between;
    + align-items: center;
    + padding: 1rem;
    + max-width: 1200px;
    + margin: 0 auto;
    + }
    +
    + .nav-content h1 {
    + color: white;
    + font-size: 1.5rem;
    + }
    +
    + .nav-links {
    + display: flex;
    + list-style: none;
    + gap: 2rem;
    + }
    +
    + .nav-links a {
    + color: white;
    + text-decoration: none;
    + transition: color 0.3s ease;
    + }
    +
    + .nav-links a:hover {
    + color: #b8e6ff;
    + }
    +
    + .menu-toggle {
    + display: none;
    + background: none;
    + border: none;
    + color: white;
    + font-size: 1.5rem;
    + cursor: pointer;
    + }
    +
    + /* Hero Section */
    + .hero {
    + padding: 8rem 0 4rem;
    + background: linear-gradient(135deg, var(--primary-color), var(--secondary-color));
    + color: white;
    + text-align: center;
    + }
    +
    + .hero h2 {
    + font-size: 2.5rem;
    + margin-bottom: 1rem;
    + }
    +
    + .authors {
    + font-style: italic;
    + margin-bottom: 1.5rem;
    + }
    +
    + .badge-container {
    + display: flex;
    + gap: 1rem;
    + justify-content: center;
    + flex-wrap: wrap;
    + }
    +
    + .badge {
    + background: rgba(255,255,255,0.2);
    + padding: 0.5rem 1rem;
    + border-radius: 20px;
    + font-size: 0.9rem;
    + }
    +
    + /* Paper Sections */
    + .paper-section {
    + padding: 4rem 0;
    + }
    +
    + .paper-section h3 {
    + color: var(--primary-color);
    + font-size: 2rem;
    + margin-bottom: 2rem;
    + }
    +
    + /* Methodology Grid */
    + .methodology-grid {
    + display: grid;
    + grid-template-columns: repeat(auto-fit, minmax(250px, 1fr));
    + gap: 2rem;
    + margin-top: 2rem;
    + }
    +
    + .method-card {
    + background: var(--card-background);
    + padding: 2rem;
    + border-radius: 8px;
    + box-shadow: 0 4px 6px rgba(0,0,0,0.1);
    + transition: transform 0.3s ease;
    + }
    +
    + .method-card:hover {
    + transform: translateY(-5px);
    + }
    +
    + /* Results Visualization */
    + .results-visualization {
    + display: flex;
    + justify-content: center;
    + gap: 3rem;
    + flex-wrap: wrap;
    + }
    +
    + .chart {
    + text-align: center;
    + width: 200px;
    + }
    +
    + .chart-bar {
    + height: 200px;
    + width: 30px;
    + margin: 0 auto 1rem;
    + background: #ddd;
    + position: relative;
    + border-radius: 15px;
    + overflow: hidden;
    + }
    +
    + .chart-bar::after {
    + content: '';
    + position: absolute;
    + bottom: 0;
    + left: 0;
    + width: 100%;
    + background: var(--accent-color);
    + transition: height 1.5s ease-out;
    + }
    +
    + /* Footer */
    + footer {
    + background: var(--secondary-color);
    + color: white;
    + padding: 2rem 0;
    + text-align: center;
    + }
    +
    + /* Responsive Design */
    + @media (max-width: 768px) {
    + .menu-toggle {
    + display: block;
    + }
    +
    + .nav-links {
    + display: none;
    + position: absolute;
    + top: 100%;
    + left: 0;
    + right: 0;
    + background: var(--primary-color);
    + flex-direction: column;
    + padding: 1rem;
    + gap: 1rem;
    + }
    +
    + .nav-links.active {
    + display: flex;
    + }
    +
    + .hero h2 {
    + font-size: 2rem;
    + }
    + }