Skip to content

Zero to Hero Series

Introduction

This repository serves as a structured documentation hub for my journey in understanding and implementing artificial intelligence and machine learning concepts, with a primary focus on building a GPT model. The content is heavily based on Andrej Karpathy's Neural Networks: Zero to Hero series and aims to provide well-organized insights, implementations, and notes for anyone following a similar path.

This site is designed for:

  • Learners and developers interested in AI and deep learning.

  • Anyone following Karpathy's Zero to Hero series.

  • The open-source community looking for structured references and implementations.

 

Series Thumbnail

 

Documentation structure

The learning resources are divided into three primary sections: Language Models, Transformer Models and General MISC.

I. Language Modeling Framework

Topic Implementation of Lecture Link Documentation Status
Backpropagation Micrograd Read notes Done
Language Model-1 Makemore (Part 1) Read notes Done
Language Model-2 Makemore (Part 2) Read notes Done
Language Model-3 Makemore (Part 3) Read notes Done
Language Model-4 Makemore (Part 4) Read notes Done
Language Model-5 Makemore (Part 5) Read notes Done

II. Transformer based Modeling Framework - GPT

Topic Implementation of Lecture Link Documentation Status
Transformer Model-1 (GPT) Build GPT Read notes Done
Transformer Model-2 (GPT) Reproducing GPT-2 Read notes Ongoing
Tokenizers Build GPT Tokenizer Read notes -

III. General MISC

Topic Lecture Link Documentation Status
Intro to LLMs Read notes Done
Deep Dive to LLMs Read notes Done

 

Reflections and Usage Guidelines

Note from the author

  • This repository is a personal knowledge base for revision, experimentation, and sharing insights.

  • If you find it useful, or spot any errors or improvements, feel free to reach out.

  • You are welcome to reference this material for your own learning journey!

Ethical Usage

This content is based on open-source educational materials (Thanks to Andrej Karpathy). While you are encouraged to learn from and contribute to this resource, please do not copy or repurpose it for commercial use. Respect the effort behind this work, and let's continue fostering an open and ethical AI learning community!