# Cross-Architecture Model Diffing with Crosscoders: Unsupervised Discovery of Differences Between LLMs

arXiv: https://arxiv.org/abs/2602.11729 · HTML: https://mlsanity.com/paper/2602.11729/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Feb 2026.

## Built on by (most cited first, 12 of 12)

- [Delta-Crosscoder: Robust Crosscoder Model Diffing in Narrow Fine-Tuning Regimes](https://mlsanity.com/paper/2603.04426/) (2026)
- [Agent trajectories as programs: fingerprinting and programming coding-agent behavior](https://mlsanity.com/paper/2606.16988/) (2026)
- [Transcoder Adapters for Reasoning-Model Diffing](https://mlsanity.com/paper/2602.20904/) (2026)
- [WriteSAE: Sparse Autoencoders for Recurrent State](https://mlsanity.com/paper/2605.12770/) (2026)
- [Actionable Activation Directions for Detecting and Mitigating Emergent Misalignment Across Language Model Families](https://mlsanity.com/paper/2606.20225/) (2026)
- [Localizing RL-Induced Tool Use to a Single Crosscoder Feature](https://mlsanity.com/paper/2606.26474/) (2026)
- [Capsule Lens: Locating and Tracking Concept Geometry in Model Representations](https://mlsanity.com/paper/2609.05575/) (2026)
- [Understanding On-Policy Distillation: A Mechanistic Interpretability Perspective via Sparse Crosscoders](https://mlsanity.com/paper/2609.35210/) (2026)
- [Exemplar Partitioning for Mechanistic Interpretability](https://mlsanity.com/paper/2605.14347/) (2026)
- [ChronoLens: Measuring Language Change Across Time, Languages, and Linguistic Levels](https://mlsanity.com/paper/2608.03507/) (2026)
- [SharedSAE: One Feature Dictionary Across Language Models](https://mlsanity.com/paper/2609.04344/) (2026)
- [Locating and Steering Refusal Beyond Attention](https://mlsanity.com/paper/2609.04721/) (2026)

## Abstract (arXiv)

Model diffing, the process of comparing models' internal representations to identify their differences, is a promising approach for uncovering safety-critical behaviors in new models. However, its application has so far been primarily focused on comparing a base model with its finetune. Since new LLM releases are often novel architectures, cross-architecture methods are essential to make model diffing widely applicable. Crosscoders are one solution capable of cross-architecture model diffing but have only ever been applied to base vs finetune comparisons. We provide the first application of crosscoders to cross-architecture model diffing and introduce Dedicated Feature Crosscoders (DFCs), an architectural modification designed to better isolate features unique to one model. Using this technique, we find in an unsupervised fashion features including Chinese Communist Party alignment in Qwen3-8B and Deepseek-R1-0528-Qwen3-8B, American exceptionalism in Llama3.1-8B-Instruct, and a copyright refusal mechanism in GPT-OSS-20B. Together, our results work towards establishing cross-architecture crosscoder model diffing as an effective method for identifying meaningful behavioral differences between AI models.