Learning Cross-Model Activation Alignments with Explicit Many-to-Many Layer Maps

Abstract:LLMs are released at a rapid pace, raising a natural question: how do two independently trained models relate, both in which layers correspond and in how features transform between them? We study this by learning an activation alignment, a map from a source model's layerwise activations to a target's. Our method, MATCHA, factors this map into a layer map, whose output is an explicit target-by-source matrix that can be extracted and inspected, and a layer-shared feature map between hidden spaces. Most of prior work fixes the layer correspondence in advance, pairing layers at roughly the same relative depth; in contrast, we learn both factors jointly from prompts. Across 42 pairs of seven models spanning three different families, MATCHA reconstructs the target's activations more faithfully and improves retrieval-based metrics substantially, w.r.t. previous approaches. The recovered maps are broadly monotone in depth but, in contrast with most previous approaches, are consistently many-to-many: each target layer draws on a band of source layers. Our alignments also enable transfer of activation-space interventions, allowing steering vectors and probes developed for one model to transfer to another.

Machine Learning (cs.LG)

arXiv:2610.09058 [cs.LG]

(or arXiv:2610.09058v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2610.09058 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.