cs.AISep 4, 2026

Technical Manual for a Toolkit for Measuring Contextual Individuation in Transformer Language Models

Authors: José Luciano Verçosa Marques, Frederico Jorge Heitmann, Daniel Omar Perez, Marcelo Vinicius de Paula, Tárcio André dos Santos Barros

Organizations: Center for Electric Mobility Research (CEMOBE) / Power Electronics Laboratories (LEPO), University of Campinas (Unicamp) · Institute of Computing (IC), University of Campinas (Unicamp) · Center for Logic, Epistemology and History of Science (CLE), University of Campinas (Unicamp)

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

A transformer language model assigns a single, context-independent vector to a word type at its embedding layer, yet is widely believed to individuate that word's occurrences by context in its later layers. Testing this belief cleanly requires a construct that holds the word form fixed while its context and intended sense vary in a controlled, labeled way. This manual documents an open toolkit built around such a construct, which we call a bridge form: a single written word that recurs, unchanged, across two or more subject domains with a different sense in each. We describe, and justify, every stage of the pipeline: the declarative specification of bridge forms and their source domains, corpus acquisition from Wikipedia, occurrence localization, layer-wise representation extraction, a domain-pairwise silhouette measurement of separation in the model's representation space, and a paired visualization protocol. Each design choice is presented together with the methodological failure mode it is meant to avoid (sense contamination from overly broad category labels, the multi-group bias of the silhouette coefficient, subword-tokenization misalignment, and axis-comparability artifacts in dimensionality-reduced plots, among others). This manuscript is a methodological and implementation reference: it does not report or interpret empirical outcomes of running the toolkit on any particular model or bridge-form set. The toolkit, its full source, and the corpora used to exercise it are archived separately (Section 9) under a persistent identifier, and are intended to be cited as an instrument by studies that use it to produce and interpret empirical results.

Figures & tables

Figure 1: The six pipeline stages, top to bottom; each stage’s output is the next stage’s input. The comment beside each box summarizes what happens at that stage and why; full detail is in the correspondingly numbered subsection of this section.
Figure 2: Two examples of the toolkit’s standard output: the paired PCA scatter (embedding layer, left; final layer, center) and the pairwise silhouette-by-layer curves (right). These two runs are shown only to illustrate the output format described in this section: panel layout, marker/color encoding, and axis conventions. The second example also happens to illustrate that the three pairwise curves in a panel need not coincide: two of them run close together across the layer range while the third stays visibly apart, which is the kind of pattern this display is built to make legible. Consistent with the scope stated in Section 1 , no explanation for why any particular pair behaves as it does is offered here; the specific pattern each panel happens to show is not reported, interpreted, or generalized as a finding of this manuscript.
ModelArchitecture familyContext directionReference
bert-base-uncasedEncoderBidirectional( Devlin et al., 2019 )
roberta-baseEncoderBidirectional( Liu et al., 2019 )
distilbert-base-uncasedEncoder (distilled)Bidirectional( Sanh et al., 2019 )
microsoft/deberta-baseEncoderBidirectional( He et al., 2021 )
gpt2DecoderCausal( Radford et al., 2019 )
EleutherAI/pythia-410mDecoderCausal( Biderman et al., 2023 )
Table 1: Model families the toolkit’s interface has been designed and checked against. The pipeline processes one model per run; comparing architectures means re-running the full pipeline with a different selection.
More in this topic

Representation Geometry in Language Models

Representation geometry in language models studies how embeddings and hidden states organize concepts, computation, and behavior.

See all 421 papers

Explore similar work

CardsList
Jun 23, 2026

Transformer-Based Language Models Across Domain Verticals: Architectures, Applications and Critical Assessment

Transformer-based language models have become the default substrate for natural language processing and the pace of new releases has made it hard for practitioners to separate durable ideas from the noise of incremental announcements. This review works at two levels. At the level of mechanism, we organise the main transformer families into a working taxonomy, covering encoder-only, decoder-only, encoder-decoder, long-context, permutation-based, and generator-discriminator variants. We then extend the discussion to post-2023 developments that changed the picture in practice: instruction tuning, reinforcement learning from human feedback, direct preference optimisation, mixture-of-experts scaling, retrieval augmentation and the current flagship model families from OpenAI, Anthropic, Google, Meta, Mistral and DeepSeek. At the level of use, we survey deployments across healthcare, finance, legal, education, customer service, creative writing and scientific work. Based on this we link each to the specific capabilities that make a transformer the appropriate tool. The contribution of this paper is a critical assessment that is based on the survey. We compare architectures on four axes that matter to deployment decisions, we quantify the trade-off between parameter count and energy cost. We also discuss how alignment methods, data provenance and benchmark saturation change what it means to call a model "state of the art". The final section lists the research questions that we think deserve more attention.

Guruprakash J, Krithika L. B

SCOPE, VIT-AP University, Amaravathi, Andhra Pradesh, 522241, India · SCORE, VIT, Katpadi, Tamil Nadu, 632014, India

LLM EvaluationLLM AlignmentTransformer
AskPDF
Sep 29, 2025

Unveiling the Mechanisms of Multi-Hop Reasoning in Transformers via Identity Bridge

Large Language Models (LLMs) excel at multi-hop reasoning in distribution, yet fail on unseen compositions, a phenomenon known as the curse of two-hop reasoning. In this work, we argue that this phenomenon can be attributed to a missing supervision on the bridge entity. We formalize this gap by introducing identity bridge, a minimal supervision that enforces a identity mapping on bridge tokens. Under this supervision, even a one-layer transformer with uniform attention (Emb-MLP) can achieve out-of-distribution (OOD) two-hop generalization. We provide a theoretical analysis demonstrating that identity bridge induces an implicit regularization effect, leading the model to establish a direct subject-to-answer association. From an empirical perspective, the performance of standard GPT-2 models aligns closely with simple Emb--MLP models across varying levels of problem complexity. Finally, analyses of fine-tuned mainstream LLMs indicate that correct two-hop predictions consistently coincide with the establishment of a subject-to-answer relationship, extending our findings to realistic settings.

Pengxiao Lin, Zheng-An Chen, Zhi-Qin John Xu

School of Mathematical Sciences, Shanghai Jiao Tong University · Institute of Natural Sciences, MOE-LSC, Shanghai Jiao Tong University · Shanghai Seres Information Technology Co., Ltd, Shanghai 200040, China.

Transformer InterpretabilityOOD GeneralizationMulti-Hop ReasoningCompositional Generalization
AskPDF
Sep 9, 2026

Contrastive Projection: Reading Transformer Internals by Differencing Logit Lenses

Reading a transformer's internal states in token space is easy to do and hard to trust: a logit lens on a single hidden state is dominated, at intermediate layers, by the generic tokens the model would predict for almost any input. We read the difference instead. Subtracting two closely matched prompts' hidden states and projecting through the unembedding cancels the shared component and surfaces what separates them, an operation equivalent to reading a RepE/ActAdd steering vector through a logit lens. Built into a training-free tracer that reads at every position, sub-layer, and head and averages over designed baselines, it traces a compound- noun MLP->attention chain in Phi-2, confirmed there by activation patching, with the same distinction recovered across three architectures by readout and probe rather than by patching; it reads what retrieval surfaces for real versus fictional entities, and reads metaphor as a set of domain-to-domain mappings rather than a single figurativity feature. A cross-seed control marks the boundary: across five networks differing only in initialization, the same distinction surfaces as almost entirely different tokens (top-10 overlap 0.08). What a computation looks like in token space is network-specific; the distinction it draws is not

Olli Tuomi

Evident Solutions Oy

Transformer InterpretabilityLanguage Model IntrospectionTransformer Attention
AskPDF