My attempt at recreating some ML learnings as I try to learn how to build LLMs from scratch.
Initially I will be going through Karpathy's YouTube lecture series called Neural Networks: Zero to Hero. We will see how far I will get...
(also using this as an excuse to set up WSL on my Desktop PC and learning how to use GPU acceleration on it)
This is mainly here to force me to stop hoarding them in open tabs :)
- Karpathy's course & repo (linked above)
- WSL setup (linked above)
- This guide from freeCodeCamp on setting up WSL for NN applications with CUDA support
- Maximme Labonne's LLM Course for a meta-list of learning paths
- Fast.ai's Practical Deep Learning for Coders, especially its new Part II
- HuggingFace's Transformers Docs