Recent work has shown that computation in language models may be human-understandable, with successful efforts to localize and intervene on both single-unit features and input-output circuits. Here, we introduce an approach which extends causal mediation experiments to automatically identify model components responsible for performing a specific subtask by solely specifying a set of \textit{desiderata}, or causal attributes of the model components executing that subtask. As a proof of concept, we apply our method to automatically discover shared \textit{variable binding circuitry} in LLaMA-13B, which retrieves variable values for multiple arithmetic tasks. Our method successfully localizes variable binding to only 9 attention heads (of the 1.6k) and one MLP in the final token's residual stream.
@article{arxiv.2307.03637,
title = {Discovering Variable Binding Circuitry with Desiderata},
author = {Xander Davies and Max Nadeau and Nikhil Prakash and Tamar Rott Shaham and David Bau},
journal= {arXiv preprint arXiv:2307.03637},
year = {2023}
}