SUMMARY AI summary by gpt-5-mini
A concise, from-scratch GPT implementation assembled while completing the NeetCode ML course (authored by Stefano, May 8, 2026). It’s an educational codebase intended for students, learners, and engineers who want to study or reproduce transformer/GPT internals by reading and running plain Python implementations. Key features: - Full transformer/GPT stack: attention head, multi-head attention, transformer block, GPT model. - Inference optimizations: KV-cache and grouped query attention. - Multiple normalization implementations (Layer, Batch, RMS). - Tokenization and data pipeline: BPE tokenizer, character vocab, dataset loader, preprocessing utilities. - Lower-level foundations built from scratch: neurons, backprop, MLP, activations, loss, training loop. - Scripts to train and generate text (train.py, generate.py). Quick start via requirements and those scripts.
Language breakdown (by bytes)
Owner
SWE @Uber | ex-Amazon
Dates
| Created on GitHub | 2026-05-08 |
| Last push | 2026-05-09 |
| First seen here | 2026-05-09 |
| Last fetched | 2026-05-09 18:15 |
Similar repos (same language)
Reverse proxy for your ai quota from the SW platform.
dea6cat/2b-agentThis llm Agent was created based on necessity for one that could simply use local models without making them hallucinate and keep them focus
Zuboy/Carbon-Cost-Optimizer-Agentic-AIAn AI agent that decides where and when to run ML training jobs to minimize cost and carbon emissions, then launches them autonomously — exposed entirely through MCP tools.
FuchaZ/lm-studio-vision-bridge给纯文本 AI agent 装上眼睛——通过 LM Studio 本地视觉模型提供 MCP 图片识别服务
Shrutirowlo/AI-Video-AgentAI-powered video generation agent that automates script analysis, scene planning, and video creation.
Shivang0/social-visionPaste an Instagram/TikTok/YouTube/X link and Claude watches it for you — transcribes audio, reads on-screen text & visuals, explains what it's about. Local, cross-platform Claude Code plugin.
markusbegerow/claude-fuer-oeffentliche-verwaltung⚠️ Experimentelle Skill-Sammlung für Digitalisierung, Fachverfahren und Organisation der deutschen öffentlichen Verwaltung (OZG, FIM, FIT-Connect, KI-Einsatz, eIDAS, BSI-IT-Grundschutz, DSGVO) – bitte testen, Issues/PRs willkommen! Keine Rechts-/Datenschutzberatung, keine verbindliche Behördenentscheidung. Keine Bürger-/Mitarbeiterdaten im Repo.
knowledgegut/course-distiller把有權使用的工作坊/課程回放消化成套品牌色的步驟式學習 PDF；最後的重組交還人腦。A Claude Code skill by 知識包小腸.