Flexible Model Interpretability through Natural Language Model Editing
This work addresses the problem of making large language models more interpretable for researchers and practitioners, but it appears incremental as it builds on existing ideas without introducing a new paradigm.
The paper tackles the challenge of model interpretability by linking it to model editing, proposing that systematic editing of model behavior based on human concepts can enhance interpretability by identifying and manipulating relevant internal representations.
Model interpretability and model editing are crucial goals in the age of large language models. Interestingly, there exists a link between these two goals: if a method is able to systematically edit model behavior with regard to a human concept of interest, this editor method can help make internal representations more interpretable by pointing towards relevant representations and systematically manipulating them.