Safety, Interpretability & Theory of LLMs

I worked on some disjoint topics around this broad track.

• Attribution techniques for Interpretability
• Enhancing Trust in LLMs
• Transformers for Non-parametric Regression