The nnsight package enables interpreting and manipulating the internals of deep learned models.

Utility for behavioral and representational analyses of Language Models

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

Code and data for "A Systematic Assessment of Syntactic Generalization in Neural Language Models"

Code for the paper "Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Bias"