skills.rest
hochoa13
hochoa13COMMUNITY
Explore all skills by this author →
GitHub

sparse-autoencoder-training

Train and analyze Sparse Autoencoders to decompose neural network activations into interpretable features.

Updated Mar 13, 2026
Search Tags:#feature discovery#sae#language models#mechanistic interpretability#neural network analysis#sparse autoencoder
One-click install
npx skills add https://github.com/hochoa13/Asistente-Harlest --skill sparse-autoencoder-training-hochoa13
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sparse-autoencoder-training
Source: https://github.com/hochoa13/Asistente-Harlest/tree/main/skills/mlops/evaluation/saelens
Command: npx skills add https://github.com/hochoa13/Asistente-Harlest --skill sparse-autoencoder-training-hochoa13

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sae-lens, transformer-lens, torch, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the challenge of understanding complex neural network activations by decomposing them into sparse, interpretable features using Sparse Autoencoders (SAEs).

Core Features & Use Cases

  • Feature Discovery: Uncover the fundamental concepts learned by language models.
  • Mechanistic Interpretability: Analyze how models represent information and study phenomena like superposition.
  • Use Case: When investigating why a model makes a specific prediction, use this Skill to identify and analyze the SAE features that most strongly activate for that input, revealing the underlying learned concepts.

Quick Start

Use the sparse-autoencoder-training skill to load and analyze pre-trained SAEs for GPT-2 small model activations.

Frequently Asked Questions about sparse-autoencoder-training

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I decompose neural network activations into interpretable features?▼

To train Sparse Autoencoders (SAEs) for analyzing language model superposition, you need the SAELens and TransformerLens libraries, alongside PyTorch, to handle model interactions and SAE operations.

What is the best way to perform mechanistic interpretability studies on GPT-2 small?▼

To debug why a language model makes a specific prediction, use this Skill to identify and analyze the SAE features that most strongly activate for that input, revealing underlying learned concepts.

Does this Skill require SAELens and TransformerLens for feature discovery?▼

Yes, this Skill requires SAELens and TransformerLens libraries to facilitate feature discovery, as these dependencies are necessary for model interaction and executing SAE operations.

Can I analyze pre-trained SAEs for GPT-2 small model activations?▼

Yes, you can analyze pre-trained SAEs for GPT-2 small model activations to investigate why a model makes a specific prediction by identifying the SAE features that strongly activate for that input.

Related Skills

Explore More Software Engineering Skills→
SethyPagnaSethyPagna

sparse-autoencoder-training

Train and analyze sparse autoencoders on transformer language model activations.

Community
Advanced
thisismynewfmail-uithisismynewfmail-ui

sparse-autoencoder-training

Train and analyze Sparse Autoencoders on neural activations with SAELens.

Community
Advanced
cloudlinesscloudliness

sparse-autoencoder-training

Train and analyze Sparse Autoencoders with SAELens on transformer activations.

Community
Advanced
pacphipacphi
6

flow-nexus-neural

Trains and deploys neural networks in distributed E2B sandbox environments.

Community
Advanced