I am a Research Scientist at Goodfire working in interpretability research. Before this, I did MATS training phase with Neel Nanda where I worked on new techniques for model diffing (SAEs on activation differences). I also worked on a variation of crosscoders for understanding how chat behavior arises from the base model.

I am a fourth-year PhD student at New York University currently on leave of absence. My research in the PhD is on scaling laws and phase transitions of diffusion models and neural networks with Eric Vanden-Eijnden and Arthur Jacot. I obtained my B.S. in Mathematics at Stanford University, where I worked on interacting particle systems with Amir Dembo

Discovering Undesired Rare Behaviors via Model Diff Amplification
S. Aranguri, T. McGrath
Used by Anthropic to evaluate Claude Sonnet 4.5, see system card (page 95)

Phase-aware Training Schedule Simplifies Learning in Flow-Based Generative Models
S. Aranguri, F. Insulla
ICLR 2025 Deep Generative Model in Machine Learning Workshop and Frontiers in Probabilistic Inference Workshop

Mixed Dynamics In Linear Networks: Unifying the Lazy and Active Regimes
Z. Tu, S. Aranguri, A. Jacot
NeurIPS 2024

Optimizing Noise Schedules of Generative Models in High Dimensions
S. Aranguri, G. Biroli, M. Mezard, E. Vanden-Eijnden

Untangling planar graphs and curves by staying positive
S. Aranguri, H. Chang, D. Fridman
ACM-SIAM Symposium on Discrete Algorithms 2022