Disentangling concept semantics via multilingual averaging in Sparse Autoencoders

TL;DR

This paper introduces a method to disentangle concept semantics in Large Language Models by averaging concept activations across multiple languages using Sparse Autoencoders, improving interpretability and alignment with formal knowledge representations.

Contribution

It proposes a novel multilingual averaging technique to isolate concept semantics in LLMs, enhancing interpretability and alignment with ontology-based knowledge.

Findings

Conceptual averaging improves alignment with ground truth mappings.

Multilingual averaging yields more accurate concept representations.

Method enhances mechanistic interpretability of LLM internal states.

Abstract

Connecting LLMs with formal knowledge representation and reasoning is a promising approach to address their shortcomings. Embeddings and sparse autoencoders are widely used to represent textual content, but the semantics are entangled with syntactic and language-specific information. We propose a method that isolates concept semantics in Large Langue Models by averaging concept activations derived via Sparse Autoencoders. We create English text representations from OWL ontology classes, translate the English into French and Chinese and then pass these texts as prompts to the Gemma 2B LLM. Using the open source Gemma Scope suite of Sparse Autoencoders, we obtain concept activations for each class and language version. We average the different language activations to derive a conceptual average. We then correlate the conceptual averages with a ground truth mapping between ontology classes. Our results give a strong indication that the conceptual average aligns to the true relationship between classes when compared with a single language by itself. The result hints at a new technique which enables mechanistic interpretation of internal network states with higher accuracy.