Language Models Understand Numbers, at Least Partially

Large language models (LLMs) have exhibited impressive competence in various tasks, but their opaque internal mechanisms hinder their use in mathematical problems. In this paper, we study a fundamental question: whether language models understand numbers, a basic element in math. Based on an assumption that LLMs should be capable of compressing numbers in their hidden states to solve mathematical problems, we construct a synthetic dataset comprising addition problems and utilize linear probes to read out input numbers from the hidden …

basic cs.cl element hinder language language models large language large language models least llms math numbers paper question study tasks