Louka Ewington-Pitsos comments on Towards Multimodal Interpretability: Learning Sparse Interpretable Features in Vision Transformers

[Louka Ewington-Pitsos]

I couldn’t find a link to the code in the article so in case anyone else wants to try to replicate I think this is it: https://​​github.com/​​HugoFry/​​mats_sae_training_for_ViTs