开源工具库，让大模型决策过程更透明
Interpreto: An Explainability Library for Transformers
- 提供归因与概念解释两类方法，统一接口支持分类与生成任务
- 实现从激活提取到概念评分的端到端概念解释流程
- 适合研究可解释性或需要调试大模型的开发者
Interpreto 是一个开源的 Python 工具库，用于解释 HuggingFace 的语言模型，涵盖从早期 BERT 变体到大语言模型（LLMs）。它提供两类互补的解释方法：归因方法和基于概念的解释。该库通过统一 API 将最新研究成果与实际工具链结合，支持分类与文本生成任务的解释工作流。其核心优势在于实现了端到端的概念解释管道（从激活提取、概念学习、解释到评分），超越了传统的特征级归因，目前在同类工具中较为罕见。项目主页：https://github.com/FOR-sight-ai/interpreto，演示网站：https://for-sight-ai.github.io/interpreto-demo/
原文摘要 · Abstract (English)
Interpreto is an open-source Python library for interpreting HuggingFace language models, from early BERT variants to LLMs. It provides two complementary families of methods: attribution methods and concept-based explanations. The library bridges recent research and practical tooling by exposing explanation workflows through a unified API for both classification and text generation. A key differentiator is its end-to-end concept-based pipeline (from activation extraction to concept learning, interpretation, and scoring), which goes beyond feature-level attributions and is uncommon in existing libraries. See GitHub: https://github.com/FOR-sight-ai/interpreto and the demo website: https://for-sight-ai.github.io/interpreto-demo/.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。