# Word Meanings in Transformer Language Models

arXiv: https://arxiv.org/abs/2508.12863 · HTML: https://mlsanity.com/paper/2508.12863/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 18 references that are in the map.
- First posted on arXiv 18 Aug 2025.

## Builds on (most cited first, 4 of 4)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Attention Heads of Large Language Models: A Survey](https://mlsanity.com/paper/2409.03752/) (2024)
- [Transformers, Contextualism, and Polysemy](https://mlsanity.com/paper/2404.09577/) (2024)

## Built on by (most cited first, 2 of 2)

- [The Expert Strikes Back: Interpreting Mixture-of-Experts Language Models at Expert Level](https://mlsanity.com/paper/2604.02178/) (2026)
- [Don't Forget Your Embeddings: Robust Knowledge Erasure via Precise Editing of Embeddings](https://mlsanity.com/paper/2606.03695/) (2026)

## Abstract (arXiv)

We investigate how word meanings are represented in the transformer language models. Specifically, we focus on whether transformer models employ something analogous to a lexical store - where each word has an entry that contains semantic information. To do this, we extracted the token embedding space of RoBERTa-base and k-means clustered it into 200 clusters. In our first study, we then manually inspected the resultant clusters to consider whether they are sensitive to semantic information. In our second study, we tested whether the clusters are sensitive to five psycholinguistic measures: valence, concreteness, iconicity, taboo, and age of acquisition. Overall, our findings were very positive - there is a wide variety of semantic information encoded within the token embedding space. This serves to rule out certain "meaning eliminativist" hypotheses about how transformer LLMs process semantic information.