Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2609.39263

Concept Subspaces Compute Beyond the Logit Lens: A Weights-Only Test for Locating Representations Upstream of Readout

Aojie Yuan·Zhiyuan Julian Su·Haiyue Zhang·Zijian Su
View original source ↗Read source document ↗Export metadata ↓

Abstract

A concept subspace's effect on model behavior does not establish how it relates to the output readout. We introduce a two-sided geometric diagnostic that measures an extracted subspace's overlap with the dominant right-singular directions of the unembedding matrix, evaluated against output-oriented positive controls. Given an extracted basis, the raw diagnostic requires only model weights. Our testbed is the Format-Agnostic Reasoning Subspace (FARS), a ten-dimensional basis extracted from eighteen reasoning concepts expressed in six surface forms. Across nine rank-matched estimators and twenty-six models, four activation-derived concept estimators carry only 0.38--0.80% mean energy in the top-ten readout span. Final-layer PCA carries 3.56%, exceeding FARS in 25 of 26 models. A same-layer next-token control, evaluated using a fitted linear translator for depth matching, carries approximately thirteen times more energy than FARS, with separation in all 25 tested models. Re-extracting FARS on ten disjoint concepts yields 62--100% cross-format retrieval across twenty-four generative models, demonstrating transfer of the extraction procedure rather than a fixed basis. A complementary four-model, three-seed intervention study finds model-dependent source-directed effects that remain well below full-vector replacement. Together, the geometry and intervention controls distinguish concept structure from dominant readout directions while limiting claims of causal sufficiency.

Explore related subjects

cs.CLcs.LG

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Aojie Yuan, Zhiyuan Julian Su, Haiyue Zhang, Zijian Su. 2026-09-30. Concept Subspaces Compute Beyond the Logit Lens: A Weights-Only Test for Locating Representations Upstream of Readout. https://arxiv.org/abs/2609.39263

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Learning From How Humans Correct

In industry NLP application, our manually labeled data has a certain number of noisy data. We present a simple method to find the noisy data and relabel them manually, meanwhile we collect the correction information. Then we present novel method to incorporate the human correction information into deep learning model. Human know how to correct noisy data. So the correction information can be inject into deep learning model. We do the experiment on our own text classification dataset, which is manually labeled, because we need to relabel the noisy data in our dataset for our industry application. The experiment result shows that our learn-on-correction method improve the classification accuracy from 91.7% to 92.5% in test dataset. The 91.7% accuracy is trained on the corrected dataset, which improve the baseline from 83.3% to 91.7% in test dataset. The accuracy under human evaluation achieves more than 97%.

cs.CL↗
arXiv2026

MBBQ: A Dataset for Cross-Lingual Comparison of Stereotypes in Generative LLMs

Generative large language models (LLMs) have been shown to exhibit harmful biases and stereotypes. While safety fine-tuning typically takes place in English, if at all, these models are being used by speakers of many different languages. There is existing evidence that the performance of these models is inconsistent across languages and that they discriminate based on demographic factors of the user. Motivated by this, we investigate whether the social stereotypes exhibited by LLMs differ as a function of the language used to prompt them, while controlling for cultural differences and task accuracy. To this end, we present MBBQ (Multilingual Bias Benchmark for Question-answering), a carefully curated version of the English BBQ dataset extended to Dutch, Spanish, and Turkish, which measures stereotypes commonly held across these languages. We further complement MBBQ with a parallel control dataset to measure task performance on the question-answering task independently of bias. Our results based on several open-source and proprietary LLMs confirm that some non-English languages suffer from bias more than English, even when controlling for cultural shifts. Moreover, we observe significant cross-lingual differences in bias behaviour for all except the most accurate models. With the release of MBBQ, we hope to encourage further research on bias in multilingual settings. The dataset and code are available at https://github.com/Veranep/MBBQ.

cs.CL↗
arXiv2026

LLMR: Knowledge Distillation with a Large Language Model-Induced Reward

Large language models have become increasingly popular and demonstrated remarkable performance in various natural language processing (NLP) tasks. However, these models are typically computationally expensive and difficult to be deployed in resource-constrained environments. In this paper, we propose LLMR, a novel knowledge distillation (KD) method based on a reward function induced from large language models. We conducted experiments on multiple datasets in the dialogue generation and summarization tasks. Empirical results demonstrate that our LLMR approach consistently outperforms traditional KD methods in different tasks and datasets.

cs.CL↗