1 repository
Mechanisms to interrupt or halt the forward pass of a neural network at specific layers.
Distinct from Execution Halting Mechanisms: The candidates focus on training or script halting, whereas this is specifically about interrupting neural network forward passes for analysis.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Execution Flow Control. Refine with filters or upvote what's useful.
TransformerLens 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
Allows interrupting the model forward pass at a specific layer to extract and analyze the resulting activation tensors.