# What, Where, and How: Disentangling the Roles of Task, Language, and Model in Code Model Representations

arXiv: https://arxiv.org/abs/2607.21491 · HTML: https://mlsanity.com/paper/2607.21491/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 28 references that are in the map.
- First posted on arXiv 23 Jul 2026.

## Builds on (most cited first, 23 of 23)

- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Similarity of Neural Network Representations Revisited](https://mlsanity.com/paper/1905.00414/) (2019)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [BERT Rediscovers the Classical NLP Pipeline](https://mlsanity.com/paper/1905.05950/) (2019)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [Towards Automated Circuit Discovery for Mechanistic Interpretability](https://mlsanity.com/paper/2304.14997/) (2023)
- [The Platonic Representation Hypothesis](https://mlsanity.com/paper/2405.07987/) (2024)
- [Finding Neurons in a Haystack: Case Studies with Sparse Probing](https://mlsanity.com/paper/2305.01610/) (2023)
- [How to use and interpret activation patching](https://mlsanity.com/paper/2404.15255/) (2024)
- [Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet](https://mlsanity.com/paper/2605.29358/) (2026)
- [Emerging Cross-lingual Structure in Pretrained Language Models](https://mlsanity.com/paper/1911.01464/) (2019)
- [A Toy Model of Universality: Reverse Engineering How Networks Learn Group Operations](https://mlsanity.com/paper/2302.03025/) (2023)
- [When Being Unseen from mBERT is just the Beginning: Handling New Languages With Multilingual Language Models](https://mlsanity.com/paper/2010.12858/) (2020)
- [Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models](https://mlsanity.com/paper/2402.16438/) (2024)
- [Universal Neurons in GPT2 Language Models](https://mlsanity.com/paper/2401.12181/) (2024)
- [What Do They Capture? -- A Structural Analysis of Pre-Trained Language Models for Source Code](https://mlsanity.com/paper/2202.06840/) (2022)
- [Towards Universality: Studying Mechanistic Similarity Across Language Model Architectures](https://mlsanity.com/paper/2410.06672/) (2024)
- [The Same But Different: Structural Similarities and Differences in Multilingual Language Modeling](https://mlsanity.com/paper/2410.09223/) (2024)
- [Quantifying Feature Space Universality Across Large Language Models via Sparse Autoencoders](https://mlsanity.com/paper/2410.06981/) (2024)
- [When Meanings Meet: Investigating the Emergence and Quality of Shared Concept Spaces during Multilingual Language Model Training](https://mlsanity.com/paper/2601.22851/) (2026)
- [CSP-Atlas: Concept-Specific Neural Circuits in a Sparse Python Transformer](https://mlsanity.com/paper/2605.24603/) (2026)

## Abstract (arXiv)

Do independently trained language models come to represent the same thing in the same way? We answer for code, extending a recently introduced concept-circuit extraction method to a 2x2 design -- Python and Rust crossed with Qwen2.5-Coder-7B and DeepSeek-Coder-V1-6.7B -- and measuring a complete inventory of grammatical concepts (58 Python, 57 Rust) identically in all four cells: the smallest design that separates what depends on the task, the language, and the model. The answer splits into three parts. What earns dedicated circuitry is set by the task: the models agree on which concepts receive circuits (Spearman ρ = 0.638 for Python, 0.673 for Rust, both p < 10-7). Where those circuits sit is set by the model: Qwen processes concepts in a late band (~L17-19), DeepSeek at L6-7, for both languages. How circuits grow across layers is also set by the model: Qwen gives its atomic concepts an early spike that DeepSeek does not. "Are circuits universal?" thus has no single answer: yes for What, no for Where and How -- universality is a property of representational content, not of computational organisation. None of this structure was fixed in advance. The agreement could have landed anywhere between independence and identity; it lands at ρ ≈ 0.65. Rust constructs receive 2-3x more concept-specific circuitry than their Python equivalents, in both models. Both models share neurons between the languages (6/7 and 7/7 paired constructs), DeepSeek 1.94x more than Qwen -- a direction no prior result predicts. And Qwen binds nine keywords of Rust's type-and-trait machinery into one tight neuron cluster (Jaccard 0.535 vs null 0.112, p < 0.001), a semantic dimension invisible in surface syntax. Ablation and linear probes confirm the circuits are functional. All claims are scoped to this 2x2; whether the per-model profile predicts a third model is the designed next test.