# What is Cross-Coder? Definition & Technical Mechanics

> **Source:** [SPIDITS AI AI Glossary](https://spidits.com/ai-glossary)  
> **Category:** Mechanistic Interpretability  
> **Canonical Citation:** [https://spidits.com/ai-glossary/cross-coder](https://spidits.com/ai-glossary/cross-coder)

## Definition
A Cross-Coder is a multi-layer Sparse Autoencoder (SAE) variant that reconstructs vector activations across multiple consecutive Transformer layers simultaneously, isolating disentangled features that persist, transform, or transfer through the network depth.

## Architectural Mechanics & Deep Dive
Cross-Coders are an advanced architectural extension of Sparse Autoencoders (SAEs) introduced in interpretability research. Traditional SAEs analyze activations layer by layer in isolation. Cross-Coders, by contrast, take activation vectors from multiple transformer layers simultaneously and pass them through a single shared dictionary layer. This allows researchers to isolate disentangled latent features that persist or transform across network depth, as well as perform model diffing to analyze how fine-tuning alters feature representations.

## Industry Applications & Real-World Use Cases
Cross-layer feature tracking, model diffing, and mechanistic interpretability

## Frequently Asked Questions

### What is a Cross-Coder in interpretability research?
A Cross-Coder is a Sparse Autoencoder trained to reconstruct activations across multiple neural network layers at once, mapping how conceptual features evolve across model depth.

### What is model diffing with Cross-Coders?
Model diffing compares activations between base models and fine-tuned models (or different checkpoints) to identify exactly which features were added, deleted, or altered during training.

## Related Terminology
- [autoencoder](https://spidits.com/ai-glossary/autoencoder)
- [transformer](https://spidits.com/ai-glossary/transformer)
- [deep-learning](https://spidits.com/ai-glossary/deep-learning)
- [explainable-ai](https://spidits.com/ai-glossary/explainable-ai)

---
*From the SPIDITS AI Knowledge Base (336+ terms indexed). Explore full technical definitions: [https://spidits.com/ai-glossary](https://spidits.com/ai-glossary)*