AI Steerability 360: A Toolkit for Steering Large Language Models

2026-03-08

AI Summary

Current approaches to steering large language models lack a unified and scalable framework for control and evaluation. This work proposes an open-source Python toolkit built upon the Hugging Face ecosystem, featuring a modular design centered around a four-dimensional control plane—encompassing input, structure, state, and output. The toolkit enables composable steering pipelines that support diverse techniques, including prompt modification, weight or architecture adjustment, activation intervention, and decoding control. By offering a flexible and extensible infrastructure under the Apache 2.0 license, it significantly lowers the barrier to developing and systematically evaluating steering strategies, thereby facilitating comprehensive analysis and tailored manipulation of large language model generation behaviors.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Large Vision Models

Application Category

• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

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.

Problem

• AI Steerability
• Large Language Models
• Model Control
• Steering Methods
• Prompt Modification

Innovation

• steering pipeline
• model control surfaces
• composable steering
• LLM evaluation benchmark
• open-source toolkit