The LLM Language Network: A Neuroscientific Approach for Identifying Causally Task-Relevant Units

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: AlKhamissi, Badr, Tuckute, Greta, Bosselut, Antoine, Schrimpf, Martin
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912231142195200
author AlKhamissi, Badr
Tuckute, Greta
Bosselut, Antoine
Schrimpf, Martin
author_facet AlKhamissi, Badr
Tuckute, Greta
Bosselut, Antoine
Schrimpf, Martin
contents Large language models (LLMs) exhibit remarkable capabilities on not just language tasks, but also various tasks that are not linguistic in nature, such as logical reasoning and social inference. In the human brain, neuroscience has identified a core language system that selectively and causally supports language processing. We here ask whether similar specialization for language emerges in LLMs. We identify language-selective units within 18 popular LLMs, using the same localization approach that is used in neuroscience. We then establish the causal role of these units by demonstrating that ablating LLM language-selective units -- but not random units -- leads to drastic deficits in language tasks. Correspondingly, language-selective LLM units are more aligned to brain recordings from the human language system than random units. Finally, we investigate whether our localization method extends to other cognitive domains: while we find specialized networks in some LLMs for reasoning and social capabilities, there are substantial differences among models. These findings provide functional and causal evidence for specialization in large language models, and highlight parallels with the functional organization in the brain.
format Preprint
id arxiv_https___arxiv_org_abs_2411_02280
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle The LLM Language Network: A Neuroscientific Approach for Identifying Causally Task-Relevant Units
AlKhamissi, Badr
Tuckute, Greta
Bosselut, Antoine
Schrimpf, Martin
Computation and Language
Machine Learning
Large language models (LLMs) exhibit remarkable capabilities on not just language tasks, but also various tasks that are not linguistic in nature, such as logical reasoning and social inference. In the human brain, neuroscience has identified a core language system that selectively and causally supports language processing. We here ask whether similar specialization for language emerges in LLMs. We identify language-selective units within 18 popular LLMs, using the same localization approach that is used in neuroscience. We then establish the causal role of these units by demonstrating that ablating LLM language-selective units -- but not random units -- leads to drastic deficits in language tasks. Correspondingly, language-selective LLM units are more aligned to brain recordings from the human language system than random units. Finally, we investigate whether our localization method extends to other cognitive domains: while we find specialized networks in some LLMs for reasoning and social capabilities, there are substantial differences among models. These findings provide functional and causal evidence for specialization in large language models, and highlight parallels with the functional organization in the brain.
title The LLM Language Network: A Neuroscientific Approach for Identifying Causally Task-Relevant Units
topic Computation and Language
Machine Learning
url https://arxiv.org/abs/2411.02280