1 dépôt
Awesome GitHub RepositoriesIntegration Validations
Testing the correctness of data flow between different modalities within a unified model.
Distinct from Multimodal Model Integrations: Focuses on verifying the correct processing of image inputs during forward passes, not the act of linking models.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Integration Validations. Refine with filters or upvote what's useful.
Awesome Integration Validations GitHub Repositories
Trouvez les meilleurs dépôts grâce à l'IA.Nous recherchons les dépôts les plus pertinents grâce à l'IA.
- transformerlensorg/transformerlens
TransformerLensOrg/TransformerLens
3,098Voir sur GitHubTransformerLens is a library for mechanistic interpretability research designed to reverse engineer the learned algorithms within large language models. It provides a standardized framework for wrapping diverse transformer architectures, allowing researchers to extract, manipulate, and analyze internal activations and weights through a consistent interface. The project distinguishes itself through a comprehensive system of activation hooks that can capture, patch, and ablate internal tensors during the forward pass. It includes specialized utilities for decomposing fused projections, material
Validates that multimodal models correctly process image inputs during forward passes and activation caching.
Python
