AI Steerability 360: A Toolkit for Steering Large Language Models

TL;DR

AI Steerability 360 introduces an open-source toolkit that provides a unified framework for steering large language models through various control surfaces, enabling easier development and evaluation of steering methods.

Contribution

It presents a comprehensive, extensible Python library with a common interface for multiple steering techniques, facilitating research and application of LLM control methods.

Findings

Toolkit supports multiple steering control surfaces

Enables composition of steering methods

Provides benchmarking and evaluation tools

Abstract

The AI Steerability 360 toolkit is an extensible, open-source Python library for steering LLMs. Steering abstractions are designed around four model control surfaces: input (modification of the prompt), structural (modification of the model's weights or architecture), state (modification of the model's activations and attentions), and output (modification of the decoding or generation process). Steering methods exert control on the model through a common interface, termed a steering pipeline, which additionally allows for the composition of multiple steering methods. Comprehensive evaluation and comparison of steering methods/pipelines is facilitated by use case classes (for defining tasks) and a benchmark class (for performance comparison on a given task). The functionality provided by the toolkit significantly lowers the barrier to developing and comprehensively evaluating steering methods. The toolkit is Hugging Face native and is released under an Apache 2.0 license at https://github.com/IBM/AISteer360.