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.
npx skills add zjunlp/Mechanist --skill clip-dissectgit clone --depth 1 https://github.com/zjunlp/MechanistWrote 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.
[](https://agentmods.dev/skills/zjunlp/mechanist/clip-dissect)<a href="https://agentmods.dev/skills/zjunlp/mechanist/clip-dissect"><img src="https://agentmods.dev/badge/skills/zjunlp/mechanist/clip-dissect.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00076 | $0.05526 |
| Opus 5 | $0.00038 | $0.02763 |
| Sonnet 5 | $0.00015 | $0.01105 |
| Haiku 4.5 | $0.00008 | $0.00553 |
Grade A, and why
clip-dissect 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 8d 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.
How it starts
The opening of the file, as written. The whole thing — 641 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLIP-Dissect Skill
When to Use
Activate this skill when:
- You need to automatically describe what individual neurons in a deep vision network respond to
- You are performing mechanistic interpretability or explainable AI research on CNNs or Vision Transformers
- You want to understand neuron-level representations in models like ResNet-50, ResNet-18, ViT, or custom models
- You need to compare neuron descriptions against baselines like NetDissect or MILAN
- You want to probe neural network layers using a concept set (e.g., 3k, 10k, 20k English words)
- You are working with Broden or ImageNet as a probing dataset
- You need to evaluate how well neuron descriptions predict class-level behavior in a model
- Keywords:
neuron dissection,CLIP,neural network interpretability,concept-based explanations,network dissection,probe dataset,activation analysis,feature visualization
Quick Reference
- Paper: CLIP-Dissect: Automatic Description of Neuron Representations in Deep Vision Networks — ICLR 2023 Spotlight
- Repository: https://github.com/Trustworthy-ML-Lab/CLIP-dissect
- CLIP Source: https://github.com/openai/CLIP
- Broden Dataset: Downloaded via
bash dlbroden.sh(based on NetDissect-Lite) - Concept Sets: Google 10k/20k words (https://github.com/first20hours/google-10000-english), EF 3k words (https://www.ef.edu/english-resources/english-vocabulary/top-3000-words/)
- PyTorch Install: https://pytorch.org/get-started/previous-versions/
Installation / Setup
Prerequisites
- Python 3.10
- PyTorch >= 1.12.0 (also compatible with 2.0), Torchvision >= 0.13
- CUDA-compatible GPU recommended (CPU inference supported)
Step-by-Step Installation
# Step 1: Clone the repository
git clone https://github.com/Trustworthy-ML-Lab/CLIP-dissect.git
cd CLIP-dissect
# Step 2: Install Python 3.10 (if not already installed)
# Using conda:
conda create -n clip_dissect python=3.10
conda activate clip_dissect
# Step 3: Install PyTorch and Torchvision (tested with 1.12.0, also works with 2.0)
# Visit https://pytorch.org/get-started/previous-versions/ for exact commands.
# Example for CUDA 11.3:
pip install torch==1.12.0+cu113 torchvision==0.13.0+cu113 --extra-index-url https://download.pytorch.org/whl/cu113
# Step 4: Install remaining dependencies
pip install -r requirements.txt
# Step 5: Download the Broden dataset (images only)
bash dlbroden.sh
# Step 6 (Optional): Download ResNet-18 pretrained on Places-365
bash dlzoo_example.sh
What ships with it
2 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.
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.
- 8d ago First seen · 641 lines · 76 tokens per session scan A 550ed295c1de
clip-dissect is a skill published in the GitHub repository zjunlp/Mechanist (72 stars, last pushed 12d ago), licensed MIT. It adds 76 tokens to every session and 5,526 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
nanoresearch-writing
Draft a LaTeX research paper from all previous stage outputs.
nanoresearch-experiment
Generate a Python code skeleton from an experiment blueprint.
nanoresearch-ideation
Search academic literature and generate research hypotheses.
nnsight-remote-interpretability
Provides guidance for interpreting and manipulating neural network internals using nnsight with optional NDIF remote execution. Use when needing to run interpretability experiments on massive models (70B+) without local GPU resources, or when working with any PyTorch architecture.
sparse-autoencoder-training
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.
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.