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.
- Score
- Unranked
- Evidence
- No signals yet
- Last commit
- Not known
- Listed
Install
armory install sparse-autoencoder-training --cli claudewrites the skill into.claude/skills/sparse-autoencoder-training/SKILL.mdListed as compatible
# fetches the source and writes it to:
.claude/skills/sparse-autoencoder-training/SKILL.mdNeeds the armory CLI · not on npm yet, build it from cli/ in the repository
What it is
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.
When to use it
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.
How to install / invoke
# Copy the skill into your .claude/skills/ directory
curl -sL https://raw.githubusercontent.com/davila7/claude-code-templates/main/cli-tool/components/skills/ai-research/mechanistic-interpretability-saelens/SKILL.md -o .claude/skills/mechanistic-interpretability-saelens/SKILL.md
Notes
Extracted from davila7/claude-code-templates, mechanistic-interpretability-saelens category.