Mechanistic Interpretability

This repository is my playground for Mechanistic Interpretability where I will push all my code-examples and readings which I am doing while practicing.

Reproduced Examples

• How to: Transformer Mechanistic Interpretability —with 50 lines of code or less!
  I was facing errors during plotting the plots in this example, I fixed these bugs using Matplotlib where I used my own matplotlib code. notebook

Important Libraries

• TransformerLens