transformer-lens-interpretability

transformer-lens-interpretability is a skill for Claude Code, Codex from davila7/claude-code-templates. It costs 52 tokens per session (3,056 once invoked), scanned A, original, MIT.

A guide to TransformerLens, a library for examining the internal steps of Transformer language models. It lets researchers inspect stored intermediate values, attention patterns, and how information moves through the model.

In plain words
What is it for?
Use it to reverse-engineer learned algorithms, analyse attention and model circuits, cache activations, attribute output scores, and run activation-patching experiments.
Why use it?
It makes hidden model behaviour easier to study when the final output alone does not explain how the answer was produced.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to reverse-engineer learned algorithms, analyse attention and model circuits, cache activations, attribute output scores, and run activation-patching experiments.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/davila7/claude-code-templates/mechanistic-interpretability-transformer-lens
About the project

Claude Code Templates is a command-line tool and catalogue for configuring Anthropic’s Claude Code with agents, commands, settings, hooks, integrations, skills, and project templates. Developers use it to browse and install reusable components for their coding workflows. The catalogue includes many of these Claude Code components.

davila7/claude-code-templates · 30,611 stars · on GitHub · aitmpl.com

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add davila7/claude-code-templates --skill mechanistic-interpretability-transformer-lens
Clone the repo
git clone --depth 1 https://github.com/davila7/claude-code-templates

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for transformer-lens-interpretability

README.md
[![agentmods](https://agentmods.dev/badge/skills/davila7/claude-code-templates/mechanistic-interpretability-transformer-lens/github.svg)](https://agentmods.dev/skills/davila7/claude-code-templates/mechanistic-interpretability-transformer-lens)
Your own site
<a href="https://agentmods.dev/skills/davila7/claude-code-templates/mechanistic-interpretability-transformer-lens"><img src="https://agentmods.dev/badge/skills/davila7/claude-code-templates/mechanistic-interpretability-transformer-lens/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for transformer-lens-interpretability

Your own site · 80×15
<a href="https://agentmods.dev/skills/davila7/claude-code-templates/mechanistic-interpretability-transformer-lens"><img src="https://agentmods.dev/badge/skills/davila7/claude-code-templates/mechanistic-interpretability-transformer-lens.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 52 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,056 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high YARA Match · line 41
    YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).
    Fix: Remove the malware payload or compromised file entirely. Investigate how it entered the skill and audit all other artifacts for additional indicators of compromise.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00052 $0.03056
Opus 5 $0.00026 $0.01528
Sonnet 5 $0.00010 $0.00611
Haiku 4.5 $0.00005 $0.00306

Measured 9d ago against content hash 9ac5a790a59f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

transformer-lens-interpretability scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

Copies of this mod

3 near-identical copies found in the catalogue:

cli-tool/components/skills/ai-research/mechanistic-interpretability-transformer-lens/SKILL.md · 347 lines

How it starts

The opening of the file, as written. The whole thing — 347 lines — stays where its author put it; the contents beside it link to each section on GitHub.

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 (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 flow
  • Analyze circuits (e.g., induction heads, IOI circuit)
  • Cache and inspect intermediate activations
  • Apply direct logit attribution

Consider alternatives when:

  • You need to work with non-transformer architectures → Use nnsight or pyvene
  • You want to train/analyze Sparse Autoencoders → Use SAELens
  • You need remote execution on massive models → Use nnsight with NDIF
  • You want higher-level causal intervention abstractions → Use pyvene

Installation

pip install transformer-lens

For development version:

pip install git+https://github.com/TransformerLensOrg/TransformerLens

Core Concepts

HookedTransformer

The main class that wraps transformer models with HookPoints on every activation:

from transformer_lens import HookedTransformer

# Load a model
model = HookedTransformer.from_pretrained("gpt2-small")

# For gated models (LLaMA, Mistral)
import os
os.environ["HF_TOKEN"] = "your_token"
model = HookedTransformer.from_pretrained("meta-llama/Llama-2-7b-hf")

Supported Models (50+)

Family Models
GPT-2 gpt2, gpt2-medium, gpt2-large, gpt2-xl
LLaMA llama-7b, llama-13b, llama-2-7b, llama-2-13b
EleutherAI pythia-70m to pythia-12b, gpt-neo, gpt-j-6b
Mistral mistral-7b, mixtral-8x7b
Others phi, qwen, opt, gemma

Read the full file on GitHub · 347 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 347 lines · 52 tokens per session scan A 9ac5a790a59f

Subscribe to this mod's changes

transformer-lens-interpretability is a skill published in the GitHub repository davila7/claude-code-templates (30,611 stars, last pushed today), licensed MIT. It adds 52 tokens to every session and 3,056 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other skills, from other repositories

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.

OpenLAIR/dr-claw · 52 tokens

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.

Orchestra-Research/AI-Research-SKILLs · 52 tokens

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.

liortesta/ClawdAgent · 52 tokens

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.

OpenLAIR/dr-claw-plugin-cc · 52 tokens

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.

ihatesea69/HieuNghi-AI-Skills · 52 tokens

saelens

Train sparse autoencoders to interpret model features.

NousResearch/hermes-agent · 14 tokens