🎯 Use Case Hermes Skill

sparse-autoencoder-training

Train sparse autoencoders to interpret model features.

仓库: NousResearch/hermes-agent · v1.0.1 · by Orchestra Research · MIT

Sparse AutoencodersSAEMechanistic InterpretabilityFeature DiscoverySuperposition

文件路径

optional-skills/mlops/saelens/SKILL.md