sparse-autoencoder-training
Guides users in training Sparse Autoencoders to analyze neural network activations and discover interpretable features.
Install this skill
or
93/100
Security score
The sparse-autoencoder-training skill was audited on May 23, 2026 and we found 7 security issues across 1 threat category. Review the findings below before installing.
Categories Tested
Security Issues
low line 331
External URL reference
SourceSKILL.md
| 331 | Browse pre-trained SAE features at [neuronpedia.org](https://neuronpedia.org): |
low line 364
External URL reference
SourceSKILL.md
| 364 | - [ARENA SAE Curriculum](https://www.lesswrong.com/posts/LnHowHgmrMbWtpkxx/intro-to-superposition-and-sparse-autoencoders-colab) |
low line 367
External URL reference
SourceSKILL.md
| 367 | - [Towards Monosemanticity](https://transformer-circuits.pub/2023/monosemantic-features) - Anthropic (2023) |
low line 368
External URL reference
SourceSKILL.md
| 368 | - [Scaling Monosemanticity](https://transformer-circuits.pub/2024/scaling-monosemanticity/) - Anthropic (2024) |
low line 369
External URL reference
SourceSKILL.md
| 369 | - [Sparse Autoencoders Find Highly Interpretable Features](https://arxiv.org/abs/2309.08600) - Cunningham et al. (ICLR 2024) |
low line 372
External URL reference
SourceSKILL.md
| 372 | - [SAELens Docs](https://jbloomaus.github.io/SAELens/) |
low line 373
External URL reference
SourceSKILL.md
| 373 | - [Neuronpedia](https://neuronpedia.org) - Feature browser |
Scanned on May 23, 2026
View Security DashboardGitHub Stars 185.0K
Rate this skill
Categorydata analytics
UpdatedJune 10, 2026
NousResearch/hermes-agent