Code associated with the paper "Evaluating the Robustness of Analogical Reasoning in Large Language Models" [https://arxiv.org/abs/2411.14215]
Code and data for each experiment -- letterstring, digit matrix, and story analogies -- is located in individual directories. Each directory has its own README with details on how to run the code.
Code for letterstring and digit matrix experiments is modified from [https://github.com/taylorwwebb/emergent_analogies_LLM/tree/main]