Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Sign in
Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

A page, a video or an image online.

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

Analyzing model activations with SAELens v1.0.0

Use this resource to train or load sparse autoencoders to decompose a model's activations

New

  • 367 downloads
  • 106 likes
  • 199 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install sparse-autoencoders-interpretability

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent guides you through installing SAELens, loading a pretrained sparse autoencoder, extracting a model's activations and encoding them into interpretable features. It also covers training new autoencoders.

Mostly picked by

Updated November 2, 2026

Card for the “Analyzing model activations with SAELens” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • SKILL.md: installation, SAE loading and activation encoding
  • references/api.md, tutorials.md and README.md: API, tutorials and getting started

Good fit for

  • You look for interpretable features in a neural network
  • You study superposition and neuron polysemanticity

Not for

    Tags

    • interpretability
    • autoencoders
    • activations
    • research

    About

    You say

    “Load a sparse autoencoder for GPT-2 and list the features that fire on this text.”

    Card for the “Analyzing model activations with SAELens” skill

    Skill published by Orchestra Research under the MIT license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to train or load sparse autoencoders to decompose a model's activations
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/api.md7.1 KB
    • references/README.md2.2 KB
    • references/tutorials.md9.5 KB
    • SKILL.md12.8 KB

    Origin and licence

    Listed unmodified

    Author
    Orchestra Research
    Source
    github.com/orchestra-research/ai-research-skills/tree/773a52944ba4747a18bd4ae9ade53fff041adcbc/04-mechanistic-interpretability/saelens
    Licence
    MIT

    Notice

    « sparse-autoencoder-training » (Orchestra Research, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.