The paper introduces LLM-Microscope, a toolkit for measuring how Large Language Models encode contextual information at the token level. It shows that seemingly minor tokens—such as determiners, stopwords, and punctuation—carry surprisingly high contextual weight, and removing them degrades performance on benchmarks like MMLU and BABILong-4k. The study also finds a strong link between contextualization and linearity, indicating that the transformation between layers can be approximated by a single linear mapping when tokens are well contextualized.
By Anton Razzhigaev, Matvey Mikhalchuk, Temurbek Rahmatullaev, Elizaveta Goncharova, Polina Druzhinina, Ivan Oseledets, Andrey Kuznetsov
The paper introduces token‑wise incremental trajectories, a method that tracks how contextual word embeddings evolve as each word is added to a sentence. Using garden‑path sentences, the authors show that these trajectories capture known processing disruptions and can distinguish ambiguous sentences from disambiguated controls. They propose metrics for measuring representational displacement and find that ambiguity signals are present not only in sentence‑level CLS vectors but also in individual token embeddings, suggesting distributed meaning construction across scales.
By Grayson Wycliffe Storer, Julia Witte Zimmerman
arXiv:2607. 04525v1 Announce Type: cross Abstract: How concepts are represented in neural networks is a fundamental question in machine learning.
By Zhimin Hu, Lanhao Niu, Sashank Varma
arXiv:2608.30315v1 Announce Type: new
Abstract: Token embeddings are the basic representational units that connect discrete tokens with continuous computation in language models. Although modern lang...
By Junjie Yao, Liangkai Hang, Zhi-Qin John Xu
How concepts are represented in neural networks is a fundamental question in machine learning. The dominant view treats concept representations as stationary geometric objects.
Neural Collapse predicts that balanced one-hot classification pushes model representations to be equally far from each other; a symmetric configuration that depends only on the output label and ignores any semantic similarity in the inputs. This creates a puzzle: next-token prediction language models are trained predominantly (as context length increases) with one-hot labels: the same context is very unlikely to appear twice in training with different labels.
arXiv:2605.24585v3 Announce Type: replace
Abstract: Neural language models are typically trained on next-token prediction, although linguistic structure spans multiple temporal scales. Successor repr...
By Mathis Immertreu, Achim Schilling, Thomas Kinfe, Patrick Krauss
The paper investigates how large language models encode and use relational information among tokens across transformer layers. By analyzing activations from prompts that require inferring relationships among three cyclic tokens (months, hours, weekdays, musical notes), the authors find a consistent layerwise progression: intermediate layers capture pairwise relationships, while later layers encode the full three‑token relationship to predict the next token. They also identify geometrically structured token relationships that do not influence prediction, and show that constraining models to use only causally relevant joint representations improves next‑token accuracy.
By Gurbir Arora, Toni J. B. Liu, Jiajun Bao, Rapha\"el Sarfati, Christopher J. Earls
arXiv:2606. 26749v1 Announce Type: new Abstract: Neural Collapse predicts that balanced one-hot classification pushes model representations to be equally far from each other; a symmetric configuration that depends only on the output label and ignores any semantic similarity in the inputs.
By Yize Zhao, Isabel Papadimitriou, Christos Thrampoulidis
The paper investigates the intrinsic dimension (ID) of large language model (LLM) representations as an indicator of linguistic complexity. By comparing ID across model layers for coordination vs. subordination, right‑branching vs. center‑embedding, and unambiguous vs. ambiguous attachment, the authors find consistent ID differences that align with established complexity contrasts. Experiments across six LLMs, including representational similarity and layer pruning analyses, confirm that more complex phenomena produce higher ID profiles, with peaks occurring at different layers for each contrast.
By Marco Baroni, Emily Cheng, Iria de-Dios-Flores, Francesca Franzon
The study investigates whether pretrained transformer models encode functional words—such as pronouns and adverbs—in a way that mirrors human usage. By comparing embeddings of nouns with those of their functional counterparts in both isolated and parallel sentences, the authors find that functional words occupy a central yet distinct position in embedding space and that parallel lexicalized and functional sentences reside in different subspaces. Experiments show that only a mixed training set of functional and lexicalized sentences reveals shared syntactic and semantic structure, whereas training on either type alone fails to capture this parallelism.
By Giuseppe Samo, Vivi Nastase, Paola Merlo
arXiv:2610.07168v1 Announce Type: new
Abstract: Representations of translated sentences are similar in the inner layers of multilingual language models -- an observation connected to the platonic rep...
By Darshil Doshi, Wenjie Zhou, Corinna Elena Wegner, Daniel J. Korchinski, Santiago Acevedo, Matthieu Wyart