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 agentmods add skills/zjunlp/mechanist/saenpx skills add zjunlp/Mechanist --skill saegit 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/sae)<a href="https://agentmods.dev/skills/zjunlp/mechanist/sae"><img src="https://agentmods.dev/badge/skills/zjunlp/mechanist/sae.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00063 | $0.00773 |
| Opus 5 | $0.00032 | $0.00387 |
| Sonnet 5 | $0.00013 | $0.00155 |
| Haiku 4.5 | $0.00006 | $0.00077 |
Grade A, and why
saelens 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 5d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SAELens — Sparse Autoencoders for Language Models
When to Use
Activate this skill when:
- Training sparse autoencoders (SAEs) on language model activations
- Loading and analyzing pre-trained SAEs from Neuronpedia or HuggingFace
- Performing mechanistic interpretability research on transformer models
- Investigating neural network features using SAE decompositions
- Generating SAE feature dashboards with SAE-Vis
- Hooking SAEs into transformer forward passes (HookedSAETransformer)
- Evaluating SAE quality (variance explained, L0 sparsity, reconstruction loss)
- Running cache activation pipelines for large-scale SAE training
- Analyzing logit lens features through SAE decomposition
- Working with TopK, Gated, or Standard SAE architectures
Keywords: sparse autoencoder, SAE, mechanistic interpretability, TransformerLens, feature analysis, neural network features, activation patching, SAE training, HookedSAETransformer, Neuronpedia, dictionary learning, GPT-2, language model interpretability
Installation / Setup
Prerequisites
- Python 3.10+
- PyTorch (CUDA recommended for training)
Install from PyPI (recommended)
pip install sae-lens
Install from source (development)
git clone https://github.com/decoderesearch/SAELens.git
cd SAELens
pip install -e ".[dev]"
Quick Reference
- Documentation: https://decoderesearch.github.io/SAELens/
- PyPI Package: https://pypi.org/project/sae-lens/
- GitHub: https://github.com/decoderesearch/SAELens
- Pre-trained SAEs list: https://decoderesearch.github.io/SAELens/latest/pretrained_saes/
- Migration guide (v6): https://decoderesearch.github.io/SAELens/latest/migrating/
- Neuronpedia: https://www.neuronpedia.org/
- SAE-Vis Library: https://github.com/callummcdougall/sae_vis
- SAEBench: https://github.com/adamkarvonen/SAEBench
- Colab Tutorials:
What ships with it
1 file 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.
- 5d ago First seen · 62 lines · 63 tokens per session scan A 49ae167519c4
saelens is a skill published in the GitHub repository zjunlp/Mechanist (55 stars, last pushed 9d ago), licensed MIT. It adds 63 tokens to every session and 773 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 2026-08-30.
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.