# Qwen-Scope: Turning Sparse Features into Development Tools for Large Language Models

arXiv: https://arxiv.org/abs/2605.11887 · HTML: https://mlsanity.com/paper/2605.11887/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- Placed in SAE features and their applications, one of 13 directions in Mechanistic interpretability.
- Builds on 59 of its 68 references that are in the map.
- First posted on arXiv 12 May 2026.

## Placed in

- Mechanistic interpretability / [SAE features and their applications](https://mlsanity.com/mechinterp/sae-features-and-their-applications/)

## Builds on (most cited first, 30 of 59)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark](https://mlsanity.com/paper/2406.01574/) (2024)
- [Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://mlsanity.com/paper/2210.09261/) (2022)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [Instruction-Following Evaluation for Large Language Models](https://mlsanity.com/paper/2311.07911/) (2023)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [OpenAI GPT-5 System Card](https://mlsanity.com/paper/2601.03267/) (2025)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [The Linear Representation Hypothesis and the Geometry of Large Language Models](https://mlsanity.com/paper/2311.03658/) (2023)
- [C-Eval: A Multi-Level Multi-Discipline Chinese Evaluation Suite for Foundation Models](https://mlsanity.com/paper/2305.08322/) (2023)
- [Scaling and evaluating sparse autoencoders](https://mlsanity.com/paper/2406.04093/) (2024)
- [Mechanistic Interpretability for AI Safety -- A Review](https://mlsanity.com/paper/2404.14082/) (2024)
- [Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2](https://mlsanity.com/paper/2408.05147/) (2024)
- [Emergent Linear Representations in World Models of Self-Supervised Sequence Models](https://mlsanity.com/paper/2309.00941/) (2023)
- [WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models](https://mlsanity.com/paper/2406.18510/) (2024)
- [Transcoders Find Interpretable LLM Feature Circuits](https://mlsanity.com/paper/2406.11944/) (2024)
- [Are We Done with MMLU?](https://mlsanity.com/paper/2406.04127/) (2024)
- [TheoremQA: A Theorem-driven Question Answering dataset](https://mlsanity.com/paper/2305.12524/) (2023)
- [Open Problems in Mechanistic Interpretability](https://mlsanity.com/paper/2501.16496/) (2025)
- [Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders](https://mlsanity.com/paper/2410.20526/) (2024)

## Built on by (most cited first, 21 of 21)

- [ICA Lens: Interpreting Language Models Without Training Another Dictionary](https://mlsanity.com/paper/2606.11722/) (2026)
- [Mechanistically Eliciting Latent Behaviors in Language Models](https://mlsanity.com/paper/2606.29604/) (2026)
- [Embodied-BenchClaw: An Autonomous Multi-Agent System for Embodied Spatial Intelligence Benchmark Construction](https://mlsanity.com/paper/2606.11909/) (2026)
- [CircuitSteer: Geometrically Aligned Multi-Layer Steering via Sparse Autoencoder Circuits](https://mlsanity.com/paper/2608.05732/) (2026)
- [Sparse Readout Prism: Explaining Logit-Lens Scores in Features Instead of Tokens](https://mlsanity.com/paper/2609.01936/) (2026)
- [ObserverBench: Testing Mechanistic Estimates for Intervention and Control](https://mlsanity.com/paper/2609.03026/) (2026)
- [Task Competence Is Not Instruction Following: Evaluating Instruction-Conflicting Behavior in Small Language Models](https://mlsanity.com/paper/2607.19608/) (2026)
- [Reading Is Not Using: Retrieval, Judgment, and the Design of AI Financial Research Workflows](https://mlsanity.com/paper/2608.24842/) (2026)
- [The Platonic brain bridge hypothesis: human brain networks as an architectural prior for multimodal large language models](https://mlsanity.com/paper/2609.10947/) (2026)
- [Decodable but Misrouted: Sparse Features Uncover a Readout Gap in Vision-Language Models for Harmful Meme Detection](https://mlsanity.com/paper/2609.18860/) (2026)
- [From Concept Alignment to Causal Grounding: An Intervention Test of Chain-of-Thought Faithfulness](https://mlsanity.com/paper/2609.23065/) (2026)
- [Perplexity Cost Understates What Activation Quantisation Breaks](https://mlsanity.com/paper/2609.23125/) (2026)
- [Complementary Roles of Activation and Parametric Memory in Few-Shot Learning](https://mlsanity.com/paper/2609.28250/) (2026)
- [Knowing Is Not Choosing: What Explicit Verification Adds Beyond Generative Preference](https://mlsanity.com/paper/2609.33142/) (2026)
- [Routing Without Embeddings: Fast And Interpretable Routing With Regular Expressions](https://mlsanity.com/paper/2609.34326/) (2026)
- [From Input to Output: A Flexible Agent for Dual-End Interpretation of Sparse Autoencoder Features](https://mlsanity.com/paper/2609.35367/) (2026)
- [Less Sycophancy, Stronger Refusal? Lessons for AI Safety from Mechanistic Interpretability](https://mlsanity.com/paper/2609.35544/) (2026)
- [Clinical Concept Centers in LLMs](https://mlsanity.com/paper/2610.02829/) (2026)
- [The Independence Prior of SAEs Fragments Visual Concepts](https://mlsanity.com/paper/2610.04112/) (2026)
- [Backdooring Sparse Autoencoders](https://mlsanity.com/paper/2610.06049/) (2026)
- [SFT Leaves a Reproducible Signature in Activation Geometry: Calibrating Post-Training Interpretability Against Run-to-Run Noise](https://mlsanity.com/paper/2610.17093/) (2026)

## Abstract (arXiv)

Large language models have achieved remarkable capabilities across diverse tasks, yet their internal decision-making processes remain largely opaque, limiting our ability to inspect, control, and systematically improve them. This opacity motivates a growing body of research in mechanistic interpretability, with sparse autoencoders (SAEs) emerging as one of the most promising tools for decomposing model activations into sparse, interpretable feature representations. We introduce Qwen-Scope, an open-source suite of SAEs built on the Qwen model family, comprising 14 groups of SAEs across 7 model variants from the Qwen3 and Qwen3.5 series, covering both dense and mixture-of-expert architectures. Built on top of these SAEs, we show that SAEs can go beyond post-hoc analysis to serve as practical interfaces for model development along four directions: (i) inference-time steering, where SAE feature directions control language, concepts, and preferences without modifying model weights; (ii) evaluation analysis, where activated SAE features provide a representation-level proxy for benchmark redundancy and capability coverage; (iii) data-centric workflows, where SAE features support multilingual toxicity classification and safety-oriented data synthesis; and (iv) post-training optimization, where SAE-derived signals are incorporated into supervised fine-tuning and reinforcement learning objectives to mitigate undesirable behaviors such as code-switching and repetition. Together, these results demonstrate that SAEs can serve not only as post-hoc analysis tools, but also as reusable representation-level interfaces for diagnosing, controlling, evaluating, and improving large language models. By open-sourcing Qwen-Scope, we aim to support mechanistic research and accelerate practical workflows that connect model internals to downstream behavior.