I am an assistant professor/faculty fellow (postdoc without a supervisor) at New York University, working on interpretability and model analysis for NLP. I aim to understand the capabilities and limitations of language models from a rigorous, interdisciplinary, and scientific perspective. I use theoretical and empirical methods from computer science, linguistics, cognitive science, and philosophy to describe, evaluate, and explain language model abilities. Current interests include analysis of Transformer representations, theory of the Transformer architecture, feature attribution and interpretability tools, and the analysis and mitigation of gender bias.