Vector Lab

Vector methods for vector theory.

The family

Comparative tools

Manifold Atlas

Object: output embeddings across models.

Concept distance, negation gauge, hegemony compass, silence detector, agonism test, vector logic. The comparative cartographic instrument.

LLMbench

Object: generated prose across models.

Dual-panel close reading, annotation, logprobs, probability visualisation. The hermeneutic surface instrument.

Single-model scopes

Vectorscope (alpha version)

Object: a single open-weight model.

Layer-by-layer weights, hidden states, attention, precision regimes. The anatomical instrument of the lab.

Manifoldscope (alpha version)

Object: a single manifold.

Intrinsic dimension, curvature, density, topology, and the political reading of what a geometry sediments or refuses. Measure and critique bound together.

Theoryscope (alpha version)

Object: a corpus of theoretical texts.

Renormalisation-group flow, eigendirections, fixed points, universality classes. A navigable geometry of theory space.

Theoretical Background

The vector lab tools are designed to open the “vector box” of artificial intelligence. The key argument is that there is a shift from the digital to the vector. Our tools and approaches have to correspondingly shift also. Commercial embedding APIs return sentence-level composites from separately-trained embedding models, the output of a pipeline rather than a window into the representations themselves. This is fine for some tasks. It is inadequate for critical work. If we want to know what the geometry of a particular model has sedimented, we have to work with open-weight models where the weights, activations, and token embeddings can be read at every layer. Vector Lab tools therefore privilege open-weight models wherever the question requires internal access, and treat commercial outputs as a separate object, the retrieval surface, with its own interest.

The result is a set of instruments that do things other research tools do not. Vectorscope runs the same inputs across precision regimes, FP32 through BF16, INT8, INT4, FP4, INT2, to observe how signal degrades as the medium is quantised, on the principle that the material substrate of a representation shapes what it can hold. Manifoldscope treats each manifold as a geometric object (intrinsic dimension, curvature, density) and as an ideological object (what it naturalises, suppresses, sediments), binding measure to critique so that no interpretive claim goes without its attestation. Theoryscope applies renormalisation-group and eigenvector methods to corpora of theoretical texts, asking which positions are fixed points under coarse-graining and which traditions are universality classes of one another. Manifold Atlas runs fifteen operations across multiple embedding models, turning particular vector-theoretic claims into empirically testable propositions. LLMbench sets two models’ prose outputs side by side and enables the dual-panel close reading that hermeneutic work requires.

The tools are designed to explore the new media theoretic landscape of AI, which are listed in each repository’s documentation and developed in the essays that motivate them. Critical work on large language models has, for a decade, been constrained by a tooling deficit. Either work with commercial APIs, and accept the interpretive costs, or do without tools and write essays that assert what a geometry looks like without ever seeing one. Vector Lab is an attempt to close that gap, to produce instruments that are continuous with the theoretical project rather than imported from outside it, and to make them public so that others can use, extend, or disagree with the work they do.

How the instruments relate

The five instruments work at different scales of the same object. Vectorscope explains the internal geometry of a single open-weight model, reading weights, activations, and attention at every layer. Manifoldscope opens one manifold for sustained close reading, often a manifold produced by a Vectorscope-inspected model, binding geometric measurement to critical interpretation. Manifold Atlas maps the terrain between models, comparing how different embedding models organise the same inputs. Manifoldscope and Atlas are anatomical and cartographic of the same kind of object, one entering a manifold and the other surveying across manifolds.

Theoryscope stands apart in its object, addressing the geometry of meso structures in vector space rather than the geometry of a model. It opens a reflexive possibility, comparing the eigendirections of a theoretical corpus against the model trained on that corpus. LLMbench is oriented to the hermeneutic reading of generated prose, the level at which models are usually encountered. It sits above the geometric instruments and often serves as the presentation layer for findings that begin lower down.

Further reading

The tools operationalise claims developed across the vector theory sequence on Stunlaw and in the wider research programme. The essays are the conceptual statements; the tools are the empirical instruments that test, extend, and sometimes contradict them.

The vector theory sequence (Stunlaw)

• The Vector Medium — names the vector regime as a periodising category for contemporary media.
• Vector Theory - gives the theoretisation of the move from digital to vectors.
• Generation Vector — examines the implications for a generation that is surrounded by AI.
• What Is Vector Space? — characterises vector space as a material substrate with grain and dimensionality, not a metaphor. Vectorscope is the empirical instrument that follows.
• What Is the Manifold? Coming soon! — develops the manifold as the medium through which large language models render meaning legible. Manifold Atlas and Manifoldscope follow.
• What Is Theory Space? Coming soon !— names the space of counterfactual media from which any particular manifold is selected.

Books and longer work

• Beguš, N. (2025) Artificial Humanities. University of Michigan Press.
• Berry, D. M. (2014) Critical Theory and the Digital. Bloomsbury.
• Berry, D.M. (2025) ‘Synthetic media and computational capitalism: towards a critical theory of artificial intelligence’, AI & SOCIETY, 40(7), pp. 5257–5269. Available at: https://doi.org/10.1007/s00146-025-02265-2.
• Impett, L. and Offert, F. (2026) Vector Media. University of Minnesota Press.
• Kittler, F. (1999) Gramophone, Film, Typewriter. Stanford University Press.
• McLuhan, M. (1962) The Gutenberg Galaxy: The Making of Typographic Man. University of Toronto Press.
• Pasquinelli, M. (2023) The Eye of the Master: A Social History of AI. Verso.
• Somaini, A. (2023) ‘A Theory of Latent Spaces’, Grey Room, 93.
• Steyerl, H. (2025) Medium Hot: Images in the Age of Heat. London: Verso.

Using the tools

Each repository contains full documentation, dependencies, and setup instructions. Tools are research instruments and are offered as-is under permissive licences. If Vector Lab tools support published research, please cite the specific tool and version, and cite the relevant theoretical essays where the claims being tested are developed.