Unveiling A Core Linguistic Region in Large Language Models
This work addresses the problem of understanding intelligence emergence in LLMs for researchers in AI and cognitive science, though it is incremental as it builds on analogies to brain localization.
The paper identified a core region in large language models (LLMs) that corresponds to linguistic competence, accounting for about 1% of parameters, and found that perturbations to single parameters can cause loss of this competence, while improvements in linguistic competence do not necessarily enhance knowledge levels.
Brain localization, which describes the association between specific regions of the brain and their corresponding functions, is widely accepted in the field of cognitive science as an objective fact. Today's large language models (LLMs) possess human-level linguistic competence and can execute complex tasks requiring abstract knowledge and reasoning. To deeply understand the inherent mechanisms of intelligence emergence in LLMs, this paper conducts an analogical research using brain localization as a prototype. We have discovered a core region in LLMs that corresponds to linguistic competence, accounting for approximately 1% of the total model parameters. This core region exhibits significant dimension dependency, and perturbations to even a single parameter on specific dimensions can lead to a loss of linguistic competence. Furthermore, we observe that an improvement in linguistic competence does not necessarily accompany an elevation in the model's knowledge level, which might imply the existence of regions of domain knowledge that are dissociated from the linguistic region. Overall, exploring the LLMs' functional regions provides insights into the foundation of their intelligence. In the future, we will continue to investigate knowledge regions within LLMs and the interactions between them.