Saved in:
Bibliographic Details
Main Authors: Wolfram, Christopher, Schein, Aaron
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.08775
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916886001745920
author Wolfram, Christopher
Schein, Aaron
author_facet Wolfram, Christopher
Schein, Aaron
contents How do the latent spaces used by independently-trained LLMs relate to one another? We study the nearest neighbor relationships induced by activations at different layers of 24 open-weight LLMs, and find that they 1) tend to vary from layer to layer within a model, and 2) are approximately shared between corresponding layers of different models. Claim 2 shows that these nearest neighbor relationships are not arbitrary, as they are shared across models, but Claim 1 shows that they are not "obvious" either, as there is no single set of nearest neighbor relationships that is universally shared. Together, these suggest that LLMs generate a progression of activation geometries from layer to layer, but that this entire progression is largely shared between models, stretched and squeezed to fit into different architectures.
format Preprint
id arxiv_https___arxiv_org_abs_2504_08775
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Layers at Similar Depths Generate Similar Activations Across LLM Architectures
Wolfram, Christopher
Schein, Aaron
Computation and Language
Artificial Intelligence
How do the latent spaces used by independently-trained LLMs relate to one another? We study the nearest neighbor relationships induced by activations at different layers of 24 open-weight LLMs, and find that they 1) tend to vary from layer to layer within a model, and 2) are approximately shared between corresponding layers of different models. Claim 2 shows that these nearest neighbor relationships are not arbitrary, as they are shared across models, but Claim 1 shows that they are not "obvious" either, as there is no single set of nearest neighbor relationships that is universally shared. Together, these suggest that LLMs generate a progression of activation geometries from layer to layer, but that this entire progression is largely shared between models, stretched and squeezed to fit into different architectures.
title Layers at Similar Depths Generate Similar Activations Across LLM Architectures
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2504.08775