sparse-autoencoder-training

仓库创建 2025年11月3日最近提交 1 个月前SkillHot 收录 21 天前

▸ 精选理由

适合研究超位置现象与单语义表示的可解释性工作。

这个 Skill 做什么

使用 SAELens 训练与分析稀疏自编码器以发现可解释神经特征。

用 SAELens 训练和分析 Sparse Autoencoders，把神经网络的稠密激活分解成稀疏、可解释的特征，方便研究 polysemantic 单元和 monosemantic 表征。适合做机制可解释性研究、发现可解释神经特征或分析模型内部语义时用。提供训练流程、分析工具和复现论文方法的实用脚本。

▸ 展开 SKILL.md 英文原文

Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.

研究检索稀疏自编码器可解释特征SAE通用

1.1w

Stars

817

Forks

仓库内 Skill

+704

7 日增星

安装 / 使用

给你的 Agent 一句话（通用）

帮我安装这个 skill：https://raw.githubusercontent.com/Orchestra-Research/AI-Research-SKILLs/main/04-mechanistic-interpretability/saelens/SKILL.md

或 curl 直取 SKILL.md

curl -fsSL "https://raw.githubusercontent.com/Orchestra-Research/AI-Research-SKILLs/main/04-mechanistic-interpretability/saelens/SKILL.md"

SKILL.MD 节选查看完整文件 ↗

# SAELens: Sparse Autoencoders for Mechanistic Interpretability

SAELens is the primary library for training and analyzing Sparse Autoencoders (SAEs) - a technique for decomposing polysemantic neural network activations into sparse, interpretable features. Based on Anthropic's groundbreaking research on monosemanticity.

**GitHub**: [jbloomAus/SAELens](https://github.com/jbloomAus/SAELens) (1,100+ stars)

## The Problem: Polysemanticity & Superposition

Individual neurons in neural networks are **polysemantic** - they activate in multiple, semantically distinct contexts. This happens because models use **superposition** to represent more features than they have neurons, making interpretabili

via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有