Visualizing LLM Latent Space Geometry Through Dimensionality Reduction

• URL: http://arxiv.org/abs/2511.21594v1
• Date: Wed, 26 Nov 2025 17:11:39 GMT
• Title: Visualizing LLM Latent Space Geometry Through Dimensionality Reduction
• Authors: Alex Ning, Vainateya Rangaraju,
• Abstract summary: We extract, process, and visualize latent state geometries in Transformer-based language models through dimensionality reduction.<n>We demonstrate experiments on GPT-2 and LLaMa models, where we uncover interesting geometric patterns in latent space.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) achieve state-of-the-art results across many natural language tasks, but their internal mechanisms remain difficult to interpret. In this work, we extract, process, and visualize latent state geometries in Transformer-based language models through dimensionality reduction. We capture layerwise activations at multiple points within Transformer blocks and enable systematic analysis through Principal Component Analysis (PCA) and Uniform Manifold Approximation (UMAP). We demonstrate experiments on GPT-2 and LLaMa models, where we uncover interesting geometric patterns in latent space. Notably, we identify a clear separation between attention and MLP component outputs across intermediate layers, a pattern not documented in prior work to our knowledge. We also characterize the high norm of latent states at the initial sequence position and visualize the layerwise evolution of latent states. Additionally, we demonstrate the high-dimensional helical structure of GPT-2's positional embeddings, the sequence-wise geometric patterns in LLaMa, and experiment with repeating token sequences. We aim to support systematic analysis of Transformer internals with the goal of enabling further reproducible interpretability research. We make our code available at https://github.com/Vainateya/Feature_Geometry_Visualization.

Related papers

• TangramPuzzle: Evaluating Multimodal Large Language Models with Compositional Spatial Reasoning [104.66714520975837]
  We introduce a geometry-grounded benchmark designed to evaluate compositional spatial reasoning through the lens of the classic Tangram game.<n>We propose the Tangram Construction Expression (TCE), a symbolic geometric framework that grounds tangram assemblies in exact, machine-verifiable coordinate specifications.<n>We conduct extensive evaluation experiments on advanced open-source and proprietary models, revealing an interesting insight: MLLMs tend to prioritize matching the target silhouette while neglecting geometric constraints.
  arXiv  Detail & Related papers  (2026-01-23T07:35:05Z)
• Inverting Self-Organizing Maps: A Unified Activation-Based Framework [39.146761527401424]
  We show that the activation pattern of a SOM can be inverted to recover the exact input under mild geometric conditions.<n>We introduce the Manifold-Aware Unified SOM Inversion and Control (MUSIC) update rule.<n>We validate the approach using synthetic Gaussian mixtures, the MNIST and the Faces in the Wild dataset.
  arXiv  Detail & Related papers  (2026-01-20T11:02:54Z)
• Geometry of Decision Making in Language Models [19.74354232642455]
  Large Language Models (LLMs) show strong generalization across diverse tasks, yet the internal decision-making processes behind their predictions remain opaque.<n>We study the geometry of hidden representations in LLMs through the lens of textitintrinsic dimension (ID)<n>We perform a large-scale study, with 28 open-weight transformer models and estimate ID across layers using multiple estimators.
  arXiv  Detail & Related papers  (2025-11-25T13:52:46Z)
• How Multimodal LLMs Solve Image Tasks: A Lens on Visual Grounding, Task Reasoning, and Answer Decoding [39.342366994703376]
  We introduce a probing framework to analyze how MLLMs process visual and textual inputs across layers.<n>We show that while the overall stage-wise structure remains stable across variations in visual tokenization, instruction tuning data, and pretraining corpus, the specific layer allocation to each stage shifts.
  arXiv  Detail & Related papers  (2025-08-27T21:22:01Z)
• Multimodal Language Models See Better When They Look Shallower [54.5303326937134]
  Multimodal large language models (MLLMs) typically extract visual features from the final layers of a pretrained Vision Transformer (ViT)<n>We present the first comprehensive study of visual layer selection for MLLMs, analyzing representation similarity across ViT layers.<n>We find that while deep layers excel in semantic-rich tasks like OCR, shallow and middle layers significantly outperform them on fine-grained visual tasks.
  arXiv  Detail & Related papers  (2025-04-30T09:07:10Z)
• Aligning Large Language Models and Geometric Deep Models for Protein Representation [57.59506688299817]
  Latent representation alignment is used to map embeddings from different modalities into a shared space, often aligned with the embedding space of large language models (LLMs)<n>Preliminary protein-focused large language models (MLLMs) have emerged, but they have predominantly relied on approaches lacking a fundamental understanding of optimal alignment practices across representations.<n>In this study, we explore the alignment of multimodal representations between LLMs and Geometric Deep Models (GDMs) in the protein domain.<n>Our work examines alignment factors from both model and protein perspectives, identifying challenges in current alignment methodologies and proposing strategies to improve the alignment process.
  arXiv  Detail & Related papers  (2024-11-08T04:15:08Z)
• Emergence of a High-Dimensional Abstraction Phase in Language Transformers [47.60397331657208]
  A language model (LM) is a mapping from a linguistic context to an output token.<n>We take a high-level geometric approach to its analysis, observing, across five pre-trained transformer-based LMs and three input datasets.<n>Our results suggest that a central high-dimensionality phase underlies core linguistic processing in many common LM architectures.
  arXiv  Detail & Related papers  (2024-05-24T11:49:07Z)
• N2F2: Hierarchical Scene Understanding with Nested Neural Feature Fields [112.02885337510716]
  Nested Neural Feature Fields (N2F2) is a novel approach that employs hierarchical supervision to learn a single feature field. We leverage a 2D class-agnostic segmentation model to provide semantically meaningful pixel groupings at arbitrary scales in the image space. Our approach outperforms the state-of-the-art feature field distillation methods on tasks such as open-vocabulary 3D segmentation and localization.
  arXiv  Detail & Related papers  (2024-03-16T18:50:44Z)
• The geometry of hidden representations of large transformer models [43.16765170255552]
  Large transformers are powerful architectures used for self-supervised data analysis across various data types. We show that the semantic structure of the dataset emerges from a sequence of transformations between one representation and the next. We show that the semantic information of the dataset is better expressed at the end of the first peak, and this phenomenon can be observed across many models trained on diverse datasets.
  arXiv  Detail & Related papers  (2023-02-01T07:50:26Z)
• Self-supervised Geometric Perception [96.89966337518854]
  Self-supervised geometric perception is a framework to learn a feature descriptor for correspondence matching without any ground-truth geometric model labels. We show that SGP achieves state-of-the-art performance that is on-par or superior to the supervised oracles trained using ground-truth labels.
  arXiv  Detail & Related papers  (2021-03-04T15:34:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.