1 repository
Awesome GitHub RepositoriesActivation Patching
The process of replacing internal activations with specific values to test neuron influence.
Distinct from Model Layer Patching: Distinct from Model Layer Patching: replaces tensor values (activations) rather than replacing the neural network layer implementation.
Explore 1 awesome GitHub repository matching testing & quality assurance · Activation Patching. Refine with filters or upvote what's useful.
Awesome Activation Patching GitHub Repositories
Găsește cele mai bune repo-uri cu AI.Vom căuta cele mai potrivite repository-uri folosind AI.
- transformerlensorg/transformerlens
TransformerLensOrg/TransformerLens
3,098Vezi pe 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
Replaces internal model activations with specific values during a forward pass to test the influence of specific neurons.
Python
