On the Scaling Laws of Geographical Representation in Language Models

arXiv:2402.19406v1 Announce Type: new
Abstract: Language models have long been shown to embed geographical information in their hidden representations. This line of work has recently been revisited by extending this result to Large Language Models (LLMs). In this paper, we propose to fill the gap between well-established and recent literature by observing how geographical knowledge evolves when scaling language models. We show that geographical knowledge is observable even for tiny models, and that it scales consistently as we increase the …

abstract arxiv cs.ai cs.cl embed gap hidden information language language models large language large language models laws line literature llms paper representation scaling type work