Steering Large Language Model Activations in Sparse Spaces

• URL: http://arxiv.org/abs/2503.00177v1
• Date: Fri, 28 Feb 2025 20:43:45 GMT
• Title: Steering Large Language Model Activations in Sparse Spaces
• Authors: Reza Bayat, Ali Rahimi-Kalahroudi, Mohammad Pezeshki, Sarath Chandar, Pascal Vincent,
• Abstract summary: A key challenge in AI alignment is guiding large language models (LLMs) to follow desired behaviors at test time.<n>We introduce sparse activation steering (SAS), a method that leverages sparse autoencoders (SAEs) to steer behavior in sparse spaces.
• Score: 21.55545768931058
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: A key challenge in AI alignment is guiding large language models (LLMs) to follow desired behaviors at test time. Activation steering, which modifies internal model activations during inference, offers a potential solution. However, prior work in dense activation spaces struggles with superposition, wherein multiple features become entangled, limiting interpretability and precise control. In contrast, sparse representations provide an untapped opportunity for more interpretable behavior modulation. In this work, we introduce sparse activation steering (SAS), a method that leverages sparse autoencoders (SAEs) to steer LLM behavior in sparse spaces. By isolating behavior-specific features through a contrastive prompt-pairing approach, we define a set of features that can selectively reinforce or suppress behaviors. Experiments on Gemma 2 LLMs show that SAS vectors enable nuanced behavioral modulation and finer-grained control. Furthermore, scaling SAEs improves monosemanticity of SAS vectors, suggesting more reliable and interpretable interventions.

Related papers

• LANGTRAJ: Diffusion Model and Dataset for Language-Conditioned Trajectory Simulation [94.84458417662404]
  LangTraj is a language-conditioned scene-diffusion model that simulates the joint behavior of all agents in traffic scenarios. By conditioning on natural language inputs, LangTraj provides flexible and intuitive control over interactive behaviors. LangTraj demonstrates strong performance in realism, language controllability, and language-conditioned safety-critical simulation.
  arXiv  Detail & Related papers  (2025-04-15T17:14:06Z)
• LF-Steering: Latent Feature Activation Steering for Enhancing Semantic Consistency in Large Language Models [16.37602070339033]
  Large Language Models (LLMs) often generate inconsistent responses when prompted with semantically equivalent paraphrased inputs.<n>We propose LF-Steering, a novel activation steering approach to precisely identify latent feature representations responsible for semantic inconsistency.<n>Our method maps the hidden states of the relevant transformer layer into a sparsely activated, high-dimensional feature space based on a sparse autoencoder.
  arXiv  Detail & Related papers  (2025-01-19T13:06:51Z)
• On the Expressiveness and Length Generalization of Selective State-Space Models on Regular Languages [56.22289522687125]
  Selective state-space models (SSMs) are an emerging alternative to the Transformer. We analyze their expressiveness and length generalization performance on regular language tasks. We introduce the Selective Dense State-Space Model (SD-SSM), the first selective SSM that exhibits perfect length generalization.
  arXiv  Detail & Related papers  (2024-12-26T20:53:04Z)
• Refusal in LLMs is an Affine Function [1.722461331472526]
  We propose affine concept editing (ACE) as an approach for steering language models' behavior. ACE combines affine subspace projection and activation addition to reliably control the model's refusal responses. Our experiments demonstrate that ACE consistently achieves more precise control over model behavior than existing methods.
  arXiv  Detail & Related papers  (2024-11-13T20:12:55Z)
• Semantics-Adaptive Activation Intervention for LLMs via Dynamic Steering Vectors [8.761404991620285]
  Activation intervention has emerged as an effective and economical method to modify the behavior of large language models (LLMs)<n>We propose Semantics-Adaptive Dynamic Intervention (SADI), a novel method that constructs a dynamic steering vector to intervene model activations at inference time.<n> Experimental results show that SADI outperforms established baselines by substantial margins, improving task performance without training.
  arXiv  Detail & Related papers  (2024-10-16T06:58:49Z)
• Model Surgery: Modulating LLM's Behavior Via Simple Parameter Editing [63.20133320524577]
  We show that editing a small subset of parameters can effectively modulate specific behaviors of large language models (LLMs)<n>Our approach achieves reductions of up to 90.0% in toxicity on the RealToxicityPrompts dataset and 49.2% on ToxiGen.
  arXiv  Detail & Related papers  (2024-07-11T17:52:03Z)
• Improving Dictionary Learning with Gated Sparse Autoencoders [8.3037652157611]
  Gated Sparse Autoencoder (Gated SAE) is a technique for unsupervised discovery of interpretable features in language models' (LMs) activations. In SAEs, the L1 penalty used to encourage sparsity introduces many undesirable biases, such as shrinkage. In training SAEs on LMs of up to 7B parameters, Gated SAEs solve shrinkage, and require half as many firing features to achieve comparable reconstruction fidelity.
  arXiv  Detail & Related papers  (2024-04-24T17:47:22Z)
• Contrastive Instruction Tuning [61.97704869248903]
  We propose Contrastive Instruction Tuning to maximize the similarity between semantically equivalent instruction-instance pairs. Experiments on the PromptBench benchmark show that CoIN consistently improves LLMs' robustness to unseen instructions with variations across character, word, sentence, and semantic levels by an average of +2.5% in accuracy.
  arXiv  Detail & Related papers  (2024-02-17T00:09:32Z)
• InferAligner: Inference-Time Alignment for Harmlessness through Cross-Model Guidance [56.184255657175335]
  We develop textbfInferAligner, a novel inference-time alignment method that utilizes cross-model guidance for harmlessness alignment. Experimental results show that our method can be very effectively applied to domain-specific models in finance, medicine, and mathematics. It significantly diminishes the Attack Success Rate (ASR) of both harmful instructions and jailbreak attacks, while maintaining almost unchanged performance in downstream tasks.
  arXiv  Detail & Related papers  (2024-01-20T10:41:03Z)
• Steering Llama 2 via Contrastive Activation Addition [41.54815073311959]
  Contrastive Activation Addition (CAA) is a method for steering language models by modifying their activations during forward passes. CAA accurately steers model outputs and sheds light on how high-level concepts are represented in Large Language Models (LLMs)
  arXiv  Detail & Related papers  (2023-12-09T04:40:46Z)
• Instruction Position Matters in Sequence Generation with Large Language Models [67.87516654892343]
  Large language models (LLMs) are capable of performing conditional sequence generation tasks, such as translation or summarization. We propose enhancing the instruction-following capability of LLMs by shifting the position of task instructions after the input sentences.
  arXiv  Detail & Related papers  (2023-08-23T12:36:57Z)
• Interpretability at Scale: Identifying Causal Mechanisms in Alpaca [62.65877150123775]
  We use Boundless DAS to efficiently search for interpretable causal structure in large language models while they follow instructions. Our findings mark a first step toward faithfully understanding the inner-workings of our ever-growing and most widely deployed language models.
  arXiv  Detail & Related papers  (2023-05-15T17:15:40Z)
• Diffusion-LM Improves Controllable Text Generation [80.50044830018442]
  Controlling the behavior of language models (LMs) without re-training is a major open problem in natural language generation. We develop a new non-autoregressive language model based on continuous diffusions that we call Diffusion-LM. We demonstrate successful control of Diffusion-LM for six challenging fine-grained control tasks, significantly outperforming prior work.
  arXiv  Detail & Related papers  (2022-05-27T20:12:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.