# Circuit Tracing in Autoregressive Protein Language Models

arXiv: https://arxiv.org/abs/2606.16044 · HTML: https://mlsanity.com/paper/2606.16044/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Interpreting biomedical and science models, one of 13 directions in Mechanistic interpretability.
- Builds on 11 of its 42 references that are in the map.
- First posted on arXiv 14 Jun 2026.

## Placed in

- Mechanistic interpretability / [Interpreting biomedical and science models](https://mlsanity.com/mechinterp/interpreting-biomedical-and-science-models/)

## Builds on (most cited first, 11 of 11)

- [k-Sparse Autoencoders](https://mlsanity.com/paper/1312.5663/) (2013)
- [Transcoders Find Interpretable LLM Feature Circuits](https://mlsanity.com/paper/2406.11944/) (2024)
- [Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet](https://mlsanity.com/paper/2605.29358/) (2026)
- [InterPLM: Discovering Interpretable Features in Protein Language Models via Sparse Autoencoders](https://mlsanity.com/paper/2412.12101/) (2024)
- [Interpreting and Steering Protein Language Models through Sparse Autoencoders](https://mlsanity.com/paper/2502.09135/) (2025)
- [Towards Interpretable Protein Structure Prediction with Sparse Autoencoders](https://mlsanity.com/paper/2503.08764/) (2025)
- [On Recovering Higher-order Interactions from Protein Language Models](https://mlsanity.com/paper/2405.06645/) (2024)
- [SHAP zero Explains Biological Sequence Models with Near-zero Marginal Cost for Future Queries](https://mlsanity.com/paper/2410.19236/) (2024)
- [Sparse Autoencoders for Low-N Protein Function Prediction and Design](https://mlsanity.com/paper/2508.18567/) (2025)
- [Protein Circuit Tracing via Cross-layer Transcoders](https://mlsanity.com/paper/2602.12026/) (2026)
- [Sparse Crosscoders for diffing MoEs and Dense models](https://mlsanity.com/paper/2603.05805/) (2026)

## Abstract (arXiv)

Protein language models (pLMs) can generate novel protein sequences with properties beyond those observed in nature, yet the mechanisms underlying protein generation remain poorly understood. Existing mechanistic interpretability methods based on sparse autoencoders and transcoders primarily focus on protein representation learning models and do not capture the computation required for autoregressive generation. Here, we introduce ProGenMech, a mechanistic interpretability framework for generative protein language models that extends cross-layer transcoders (CLTs) to ProGen3, a sparse Mixture-of-Experts model trained for both causal generation and span infilling. Unlike per-layer approaches, CLTs reconstruct each layer using sparse latent variables from all preceding layers, enabling faithful recovery of inter-layer generative computation. We further develop a zero-shot circuit discovery framework to identify sparse latent circuits responsible for protein generation and fitness prediction. In causal generation and zero-shot fitness estimation tasks, ProGenMech outperforms local transcoder baselines in recovering ProGen3's probability distribution and functional scoring behavior, while matching the original model's generative distribution in span infilling tasks. Moreover, the recovered circuits reveal biologically meaningful motifs and functional regions associated with conserved sequence patterns and protein fitness landscapes, establishing a foundation for interpretable and steerable protein generation.