Hyperdimensional Probe: Decoding LLM Representations via Vector Symbolic Architectures

ICLR 2026 Conference Submission12823 Authors

Keywords: Information Decoding, Vector Symbolic Architectures, Hyperdimensional Computing, LLMs, Probing, Residual Stream, Interpretability, Neurosymbolic, Neural representations, Concept extraction

TL;DR: This work combines symbolic representations and neural probing to introduce Hyperdimensional Probe, a new paradigm for decoding LLM vector space into human-interpretable features, consistently extracting meaningful concepts across models and inputs

Abstract:

Despite their capabilities, Large Language Models (LLMs) remain opaque with limited understanding of their internal representations.
Current interpretability methods, such as direct logit attribution (DLA) and sparse autoencoders (SAEs), provide restricted insight due to limitations such as the model's output vocabulary or unclear feature names.
This work introduces *Hyperdimensional Probe*, a novel paradigm for decoding information from the LLM vector space. It combines ideas from symbolic representations and neural probing to project the model's residual stream into interpretable concepts via Vector Symbolic Architectures (VSAs).
This probe combines the strengths of SAEs and conventional probes while overcoming their key limitations.
We validate our decoding paradigm with controlled input–completion tasks, probing the model’s final state before next-token prediction on inputs spanning syntactic pattern recognition, key–value associations, and abstract inference.
We further assess it in a question-answering setting, examining the state of the model both before and after text generation.
Our experiments show that our probe reliably extracts meaningful concepts across varied LLMs, embedding sizes, and input domains, also helping identify LLM failures.
Our work advances information decoding in LLM vector space, enabling extracting more informative, interpretable, and structured features from neural representations.

Primary Area: interpretability and explainable AI

Submission Number: 12823