Notes Storage

Herein I keep my notes on books I read to share.

Many notes are in Chinese

Building Large Language Models from Scratch

ISBN 9781633437166

My favorate pop science book! Less than half a month to complete.

Contents

• LLM architecture
• Text processing and tokenization
• Attention mechanism
• GPT model construction
• Self-supervised pre-training
• Classification fine-tuning
• Instruction fine-tuning

Operation Systems Three Easy Pieces

http://www.ostep.org

Interesting and clear book about OS. I like it! Notes in English

Currently reading.

Transformer Models for NLP and Computer Vision

ISBN 9781805128724

I don't like it. Author keeps mentioning a certain topic will be discussed in a given chapter.

However when we reach that chapter, only superficial talks and claims we have covered it before.

Not completed. Only broader pop-sci of NLP notes.

OpenManus Framework

https://github.com/FoundationAgents/OpenManus

A simple go through Agent Framework. Written by AI, just for fun.

Content:

• Agent system architecture
• Tool system
• ReAct pattern
• Browser, File and terminal tools

Deep Learning for Computer Vision

ISBN 9781617296192

Content:

• Deep learning fundamentals
• Advanced CNN architectures
• Generative Adversarial Networks

Buzz

AI fastfood

Notes taken while talking with AI. A shortcut to gain without pay lol.

• fun facts of AIGC
• Freud's theory walk through
• ...

---

⚠️ Disclaimer Some screenshot is from original books and used for educational purposes only. please contact me if any copyright concerns.