Stuff I'm Reading

These are some resources I am currently going through to learn about mechanistic interpretability.

• Neel Nanda’s Website
• Anthropic’s Transformer Circuit Threads
• ARENA 3.0 course

This is just some random stuff I’ve worked on (mostly for self-learning purposes).

• Implementing a Transformer from Scratch
• Reverse-engineering Induction Circuits
• Exploring Interpretability with Captum

Some other random stuff:

• Learning about Contrastive Learing