Aavishkar Gautam

Hi, I’m Aavishkar. I studied Computer Science at NYU Abu Dhabi. I am interested in Mechanistic Interpretability research. I work as an AI-Engineer at Wio Bank in Abu Dhabi.

Research

I am interested in AI safety research. I am worked on LLM Interpretability by using methods from Mechanistic Interpretability (MechInterp) and Representation Engineering (RepE). My recent work focuses on refusal mechanism in LLMs. You can find more about my research writeups in the blogs section.