Last released Feb 28, 2026
Utils and mechanistic interpretability intervensions using nnsight
Last released Feb 11, 2025
A tool for visualizing and exploring feature activations in neural language models.
Supported by