BlueDot AI Safety Fundamentals Alignment Course Project

This project was completed for the summer 2024 cohort of the BlueDot AI Safety Fundamentals Alignment Course.

Introduction to Mechanistic Interpretability

For my project, I followed Neel Nanda's walkthrough of his paper "Progress Measures for Grokking via Mechanistic Interpretability"¹

Footnotes

1. N. Nanda, L. Chan, T. Lieberum, J. Smith, and J. Steinhardt, ‘Progress measures for grokking via mechanistic interpretability’. arXiv, Oct. 19, 2023. doi: 10.48550/arXiv.2301.05217. ↩