skill

transformer-lens-interpretability

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.

Description as published by the maintainer. Source

• active

active — Most recent push to the repository was 2026-06-16.

Signals

These are separate measurements of different things. They are deliberately not combined into one score, because a popularity number that mixes website traffic with saves and stars cannot be checked or acted on.

Will this work with your setup?

Install location suggests this is meant for claude-code. The author tagged this repository "claude-skills" on GitHub. That is their statement of intent, not a test result.

We have not run this skill against a task with and without it enabled, so we cannot tell you whether it improves anything, what it costs in tokens, or whether it duplicates behaviour your harness already has. When we have run that test, the result will appear on this page with the task, the versions and the budget it used.

The skill definition lives at 04-mechanistic-interpretability/transformer-lens/SKILL.md in https://github.com/Orchestra-Research/AI-Research-SKILLs.

Where to get it

• Repository — https://github.com/Orchestra-Research/AI-Research-SKILLs

Also from orchestra-research

• orchestra-research-ai-research-skills-autoresearch — last commit 2026-06-16
  Orchestrates end-to-end autonomous AI research projects using a two-loop architecture. The inner loop runs rapid experi…
• orchestra-research-ai-research-skills-axolotl — last commit 2026-06-16
  Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimoda…
• orchestra-research-ai-research-skills-constitutional-ai — last commit 2026-06-16
  Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with sel…
• orchestra-research-ai-research-skills-distributed-llm-pretraining-torchtitan — last commit 2026-06-16
  Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when…
• orchestra-research-ai-research-skills-fine-tuning-with-trl — last commit 2026-06-16
  Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GR…
• orchestra-research-ai-research-skills-grpo-rl-training — last commit 2026-06-16
  Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training
• orchestra-research-ai-research-skills-huggingface-tokenizers — last commit 2026-06-16
  Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports…
• orchestra-research-ai-research-skills-implementing-llms-litgpt — last commit 2026-06-16
  Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mist…
• orchestra-research-ai-research-skills-llama-factory — last commit 2026-06-16
  Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal…
• orchestra-research-ai-research-skills-llamaguard — last commit 2026-06-16
  Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual co…