Sparse Autoencoders Find Highly Interpretable Features in Language Models

@Article{Cunningham2023SparseAF,
 author = {Hoagy Cunningham and Aidan Ewart and Logan Riggs and R. Huben and Lee Sharkey},
 booktitle = {International Conference on Learning Representations},
 journal = {ArXiv},
 title = {Sparse Autoencoders Find Highly Interpretable Features in Language Models},
 volume = {abs/2309.08600},
 year = {2023}
}