FlexModel: A Framework for Interpretability of Distributed Large Language Models
This work addresses the problem of accessibility for researchers in interpretability and responsible AI by simplifying interactions with distributed models, though it is incremental as it builds on existing model distribution libraries.
The authors tackled the challenge of making distributed large language models more accessible for interpretability research by introducing FlexModel, a software package that provides a streamlined interface for interacting with models across multi-GPU and multi-node configurations, bridging the gap for researchers with machine learning expertise but limited distributed computing knowledge.
With the growth of large language models, now incorporating billions of parameters, the hardware prerequisites for their training and deployment have seen a corresponding increase. Although existing tools facilitate model parallelization and distributed training, deeper model interactions, crucial for interpretability and responsible AI techniques, still demand thorough knowledge of distributed computing. This often hinders contributions from researchers with machine learning expertise but limited distributed computing background. Addressing this challenge, we present FlexModel, a software package providing a streamlined interface for engaging with models distributed across multi-GPU and multi-node configurations. The library is compatible with existing model distribution libraries and encapsulates PyTorch models. It exposes user-registerable HookFunctions to facilitate straightforward interaction with distributed model internals, bridging the gap between distributed and single-device model paradigms. Primarily, FlexModel enhances accessibility by democratizing model interactions and promotes more inclusive research in the domain of large-scale neural networks. The package is found at https://github.com/VectorInstitute/flex_model.