Understanding the Inner-workings of Language Models Through Representation Dissimilarity

Davis Brown, Charles Godfrey, Nicholas Konz, Jonathan Tu, Henry Kvinge

Submission Type: Regular Short Paper

Submission Track: Interpretability, Interactivity, and Analysis of Models for NLP

Submission Track 2: Language Modeling and Analysis of Language Models

Keywords: language models, interpretability, stitching

TL;DR: Using representation dissimilarity methods to understand generalization and scaling behavior in language models.

Abstract: As language models are applied to an increasing number of real-world applications, understanding their inner workings has become an important issue in model trust, interpretability, and transparency. In this work we show that representation dissimilarity measures, which are functions that measure the extent to which two model's internal representations differ, can be a valuable tool for gaining insight into the mechanics of language models. Among our insights are: (i) an apparent asymmetry in the internal representations of model using SoLU and GeLU activation functions, (ii) evidence that dissimilarity measures can identify and locate generalization properties of models that are invisible via in-distribution test set performance, and (iii) new evaluations of how language model features vary as width and depth are increased. Our results suggest that dissimilarity measures are a promising set of tools for shedding light on the inner workings of language models.

Submission Number: 4690