Pinned Loading

1. moe_refusal_decomp moe_refusal_decomp Public

   Code for "Decomposing Refusal Steering in Mixture-of-Experts Models"

   Python 1

2. glue_geometry glue_geometry Public

   Code for 'Exploring the Impact of a Transformer's Latent Space Geometry on Downstream Task Performance' (arXiv:2406.12159)

   Python

3. measures_of_spread measures_of_spread Public

   Code for 'Reliable Measures of Spread in High Dimensional Latent Spaces' (ICML 2023)

   Python

4. masking_and_analysis masking_and_analysis Public

   Jupyter Notebook