Skip to content

IntelLabs/blueglass

BlueGlass Logo

🔷 BlueGlass

License Hugging Face OpenSSF Scorecard Ask DeepWiki

BlueGlass is an open-source framework for interpretability and analysis of vision-language and vision-only models. It empowers researchers and practitioners to understand what deep models see and focus on by offering tools to extract, probe, and build ad hoc XAI models for internal representations.

The framework includes a custom infrastructure and lightweight modifications to third-party model libraries, enabling seamless feature extraction from the models. These extracted features are compiled into structured datasets called BlueLens, laying the foundation for advanced Explainable AI research and experimentation.


✨ Key Features

  • Benchmarking
    Seamless support for benchmarking different models with various datasets and evaluators.

  • BlueLens Dataset Extraction

    • Effortlessly generate feature datasets for training ad-hoc models by capturing model activations using a custom interceptor manager - Feature Recorder

      BlueLens Logo
  • 🧪 Ad-hoc Interpretability Models
    Leverage the BlueLens dataset to train models like:

    • Linear Probes
    • Sparse Autoencoders (SAEs)
    • Other direct interpretability techniques
      All optimized for large batch sizes and efficient performance.
  • 🧰 Generate Custom BlueLens Dataset
    Use the built-in tools to generate your own BlueLens dataset from any model.

  • 🔁 Feature Patching & Re-Integration
    Ad-hoc models can be patched back into the base model using the Feature Patcher, enabling:

    • Validation of interpretability methods
    • Performance benchmarking of modified models
  • 🚀 Streamlined Runner
    A simplified, plug-and-play design built around modular runners and interfaces for training, validating, and evaluating ad-hoc models.

  • ⚙️ Scalable Training

    • Native support for distributed training
    • Central control over precision (e.g., FP16, BF16)
  • 🔍 Interpretability & Analysis
    Built for both vision-language and vision-only models with a focus on clarity, transparency, and scalability.


📦 Coming Soon

  • 📚 Tutorials and example notebooks

🧪 Get Started

Coming soon: Installation guide, usage examples, and API reference.


📖 Detailed Documentation

🤝 Contributing

We welcome contributions! Feel free to open issues or submit pull requests to help expand the BlueGlass ecosystem.


About

No description, website, or topics provided.

Resources

License

Code of conduct

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •