Hello, I’m Favour Umejesi, a third-year Computer Science student at Grambling State University with a minor in Mathematics.

My current research focuses on the mechanistic interpretability of large language models, where I investigate how internal model components contribute to factual recall, hallucinations, and biased outputs. Under the supervision of my Deep Learning professor, Dr. Vivya Kalindindi, I am exploring techniques such as layer ablation and game-theoretic methods to better understand how information is represented and processed within LLMs.