TransformerLens Mechanistic Interpretability
技能 已验证 活跃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.
To enable researchers to deeply inspect and manipulate the internals of transformer models for understanding their learned algorithms and behavior.
功能
- Inspect and manipulate transformer internals
- Utilize HookPoints and activation caching
- Perform activation patching and causal tracing
- Analyze attention patterns and circuits
- Support for 50+ transformer architectures
使用场景
- Reverse-engineering model algorithms
- Studying attention patterns and information flow
- Performing activation patching or causal tracing experiments
- Analyzing specific circuits like induction heads
非目标
- Working with non-transformer architectures
- Training or analyzing Sparse Autoencoders directly
- Providing remote execution on massive models
- Offering higher-level causal intervention abstractions
实践
- Model Interpretability
- Transformer Analysis
- Code Research
先决条件
- Python >= 3.8
- transformer-lens >= 2.0.0
- torch >= 2.0.0
Trust
- info:Issues AttentionOpen issues (90d): 17, Closed issues (90d): 4. The closure rate is low, indicating slower response times for open issues.
安装
npx skills add davila7/claude-code-templates通过 npx 运行 Vercel skills CLI(skills.sh)— 需要本地安装 Node.js,以及至少一个兼容 skills 的智能体(Claude Code、Cursor、Codex 等)。前提是仓库遵循 agentskills.io 格式。
质量评分
已验证类似扩展
Transformer Lens Interpretability
99Provides 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.
Embedding Strategies
100Select and optimize embedding models for semantic search and RAG applications. Use when choosing embedding models, implementing chunking strategies, or optimizing embedding quality for specific domains.
Aws Cdk Development
100AWS Cloud Development Kit (CDK) 专家，用于使用 TypeScript/Python 构建云基础设施。在创建 CDK 堆栈、定义 CDK 构造、实现基础设施即代码，或当用户提及 CDK、CloudFormation、IaC、cdk synth、cdk deploy，或希望以编程方式定义 AWS 基础设施时使用。涵盖 CDK 应用结构、构造模式、堆栈组合和部署工作流。
Fit Drift Diffusion Model
100Fit cognitive drift-diffusion models (Ratcliff DDM) to reaction time and accuracy data with parameter estimation (drift rate, boundary separation, non-decision time), model comparison, and parameter recovery validation. Use when modeling binary decision-making with reaction time data, estimating cognitive parameters from experimental data, comparing sequential sampling model variants, or decomposing speed-accuracy tradeoff effects into latent cognitive components.
Ui Ux Pro Max
100UI/UX design intelligence with searchable style, palette, typography, and chart databases. Use when designing UI components, choosing colors/fonts, reviewing code for UX issues, building landing pages, or implementing responsive layouts.
Google Tts
100Convert documents and text to audio using Google Cloud Text-to-Speech. Use this skill when the user wants to: narrate a document, read aloud text, generate audio from a file, convert text to speech, create a recording of documentation or analysis, create a podcast from a document, or use Google TTS/text-to-speech. Trigger phrases: "read this aloud", "narrate this", "create a recording", "text to speech", "TTS", "convert to audio", "audio from document", "listen to this", "generate audio", "google tts", "create a podcast".