Interpreting and Steering LLMs with Mutual Information-based Explanations on Sparse Autoencoders

• URL: http://arxiv.org/abs/2502.15576v1
• Date: Fri, 21 Feb 2025 16:36:42 GMT
• Title: Interpreting and Steering LLMs with Mutual Information-based Explanations on Sparse Autoencoders
• Authors: Xuansheng Wu, Jiayi Yuan, Wenlin Yao, Xiaoming Zhai, Ninghao Liu,
• Abstract summary: Large language models (LLMs) excel at handling human queries, but they can occasionally generate flawed or unexpected responses.<n>We propose using a fixed vocabulary set for feature interpretations and designing a mutual information-based objective.<n>We propose two runtime steering strategies that adjust the learned feature activations based on their corresponding explanations.
• Score: 29.356200147371275
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Large language models (LLMs) excel at handling human queries, but they can occasionally generate flawed or unexpected responses. Understanding their internal states is crucial for understanding their successes, diagnosing their failures, and refining their capabilities. Although sparse autoencoders (SAEs) have shown promise for interpreting LLM internal representations, limited research has explored how to better explain SAE features, i.e., understanding the semantic meaning of features learned by SAE. Our theoretical analysis reveals that existing explanation methods suffer from the frequency bias issue, where they emphasize linguistic patterns over semantic concepts, while the latter is more critical to steer LLM behaviors. To address this, we propose using a fixed vocabulary set for feature interpretations and designing a mutual information-based objective, aiming to better capture the semantic meaning behind these features. We further propose two runtime steering strategies that adjust the learned feature activations based on their corresponding explanations. Empirical results show that, compared to baselines, our method provides more discourse-level explanations and effectively steers LLM behaviors to defend against jailbreak attacks. These findings highlight the value of explanations for steering LLM behaviors in downstream applications. We will release our code and data once accepted.

Related papers

• SEER: Self-Explainability Enhancement of Large Language Models' Representations [18.840860385644316]
  We propose a self-explaining method SEER to explain Large Language Models (LLMs)<n>In this paper, we propose a self-explaining method SEER, enhancing LLMs' explainability by aggregating the same concept and disentangling the different concepts in the representation space.<n>We showcase the applications of SEER on trustworthiness-related tasks, where self-explained LLMs achieve consistent improvement in explainability and performance.
  arXiv  Detail & Related papers  (2025-02-07T13:25:33Z)
• LatentQA: Teaching LLMs to Decode Activations Into Natural Language [72.87064562349742]
  We introduce LatentQA, the task of answering open-ended questions about model activations in natural language. We propose Latent Interpretation Tuning (LIT), which finetunes a decoder LLM on a dataset of activations and associated question-answer pairs. Our decoder also specifies a differentiable loss that we use to control models, such as debiasing models on stereotyped sentences and controlling the sentiment of generations.
  arXiv  Detail & Related papers  (2024-12-11T18:59:33Z)
• PromptExp: Multi-granularity Prompt Explanation of Large Language Models [16.259208045898415]
  We introduce PromptExp, a framework for multi-granularity prompt explanations by aggregating token-level insights. PromptExp supports both white-box and black-box explanations and extends explanations to higher granularity levels. We evaluate PromptExp in case studies such as sentiment analysis, showing the perturbation-based approach performs best.
  arXiv  Detail & Related papers  (2024-10-16T22:25:15Z)
• Traffic Light or Light Traffic? Investigating Phrasal Semantics in Large Language Models [41.233879429714925]
  This study critically examines the capacity of API-based large language models to comprehend phrase semantics. We assess the performance of LLMs in executing phrase semantic reasoning tasks guided by natural language instructions. We conduct detailed error analyses to interpret the limitations faced by LLMs in comprehending phrase semantics.
  arXiv  Detail & Related papers  (2024-10-03T08:44:17Z)
• Rethinking Semantic Parsing for Large Language Models: Enhancing LLM Performance with Semantic Hints [20.844061807562436]
  We propose SENSE, a novel prompting approach that embeds semantic hints within the prompt. Experiments show that SENSE consistently improves LLMs' performance across various tasks.
  arXiv  Detail & Related papers  (2024-09-22T14:35:09Z)
• FaithLM: Towards Faithful Explanations for Large Language Models [67.29893340289779]
  Large Language Models (LLMs) have become proficient in addressing complex tasks by leveraging their internal knowledge and reasoning capabilities. The black-box nature of these models complicates the task of explaining their decision-making processes. We introduce FaithLM to explain the decision of LLMs with natural language (NL) explanations.
  arXiv  Detail & Related papers  (2024-02-07T09:09:14Z)
• LLMCheckup: Conversational Examination of Large Language Models via Interpretability Tools and Self-Explanations [26.340786701393768]
  Interpretability tools that offer explanations in the form of a dialogue have demonstrated their efficacy in enhancing users' understanding. Current solutions for dialogue-based explanations, however, often require external tools and modules and are not easily transferable to tasks they were not designed for. We present an easily accessible tool that allows users to chat with any state-of-the-art large language model (LLM) about its behavior.
  arXiv  Detail & Related papers  (2024-01-23T09:11:07Z)
• Towards ASR Robust Spoken Language Understanding Through In-Context Learning With Word Confusion Networks [68.79880423713597]
  We introduce a method that utilizes the ASR system's lattice output instead of relying solely on the top hypothesis. Our in-context learning experiments, covering spoken question answering and intent classification, underline the LLM's resilience to noisy speech transcripts.
  arXiv  Detail & Related papers  (2024-01-05T17:58:10Z)
• From Language Modeling to Instruction Following: Understanding the Behavior Shift in LLMs after Instruction Tuning [63.63840740526497]
  We investigate how instruction tuning adjusts pre-trained models with a focus on intrinsic changes. The impact of instruction tuning is then studied by comparing the explanations derived from the pre-trained and instruction-tuned models. Our findings reveal three significant impacts of instruction tuning.
  arXiv  Detail & Related papers  (2023-09-30T21:16:05Z)
• Evaluating and Explaining Large Language Models for Code Using Syntactic Structures [74.93762031957883]
  This paper introduces ASTxplainer, an explainability method specific to Large Language Models for code. At its core, ASTxplainer provides an automated method for aligning token predictions with AST nodes. We perform an empirical evaluation on 12 popular LLMs for code using a curated dataset of the most popular GitHub projects.
  arXiv  Detail & Related papers  (2023-08-07T18:50:57Z)
• IERL: Interpretable Ensemble Representation Learning -- Combining CrowdSourced Knowledge and Distributed Semantic Representations [11.008412414253662]
  Large Language Models (LLMs) encode meanings of words in the form of distributed semantics. Recent studies have shown that LLMs tend to generate unintended, inconsistent, or wrong texts as outputs. We propose a novel ensemble learning method, Interpretable Ensemble Representation Learning (IERL), that systematically combines LLM and crowdsourced knowledge representations.
  arXiv  Detail & Related papers  (2023-06-24T05:02:34Z)
• Label Words are Anchors: An Information Flow Perspective for Understanding In-Context Learning [77.7070536959126]
  In-context learning (ICL) emerges as a promising capability of large language models (LLMs) In this paper, we investigate the working mechanism of ICL through an information flow lens. We introduce an anchor re-weighting method to improve ICL performance, a demonstration compression technique to expedite inference, and an analysis framework for diagnosing ICL errors in GPT2-XL.
  arXiv  Detail & Related papers  (2023-05-23T15:26:20Z)
• Shortcut Learning of Large Language Models in Natural Language Understanding [119.45683008451698]
  Large language models (LLMs) have achieved state-of-the-art performance on a series of natural language understanding tasks. They might rely on dataset bias and artifacts as shortcuts for prediction. This has significantly affected their generalizability and adversarial robustness.
  arXiv  Detail & Related papers  (2022-08-25T03:51:39Z)

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.