transformer-lens-interpretability
仓库创建 2025年11月3日最近提交 1 个月前SkillHot 收录 20 天前
▸ 精选理由
社区事实标准，接口清晰，适合深入机制逆向与实验复现。
这个 Skill 做什么
使用 TransformerLens 检查并操作 Transformer 内部激活与注意力以做可解释性研究。
用 TransformerLens 去查看和操纵 Transformer 的内部激活与注意力，适合做机制性可解释性研究。在反推模型算法、研究 attention 模式或做 activation patching 实验时使用，可以在 HookPoints 上缓存并注入激活值。特点是能细粒度定位某层或某个 attention 头的功能，便于做可重现的实验和可视化分析。
▸ 展开 SKILL.md 英文原文
Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.
1.1w
Stars
817
Forks
40
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/transformer-lens/SKILL.md或 curl 直取 SKILL.md
curl -fsSL "https://raw.githubusercontent.com/Orchestra-Research/AI-Research-SKILLs/main/04-mechanistic-interpretability/transformer-lens/SKILL.md"SKILL.MD 节选查看完整文件 ↗
# TransformerLens: Mechanistic Interpretability for Transformers TransformerLens is the de facto standard library for mechanistic interpretability research on GPT-style language models. Created by Neel Nanda and maintained by Bryce Meyer, it provides clean interfaces to inspect and manipulate model internals via HookPoints on every activation. **GitHub**: [TransformerLensOrg/TransformerLens](https://github.com/TransformerLensOrg/TransformerLens) (2,900+ stars) ## When to Use TransformerLens **Use TransformerLens when you need to:** - Reverse-engineer algorithms learned during training - Perform activation patching / causal tracing experiments - Study attention patterns and information fl
via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有