EleutherAI SAE

Tool (not an eval — tooling/reference entry)

Train sparse autoencoders/transcoders on activations.

Category

Model-internals auditing

Insurance exposure

Model failure

Publisher

EleutherAI

Publisher type

Nonprofit / research org

License

MIT

Status

Active (<90d)

Last commit

2026-10-12

Stars

742

View source on GitHub →

Wanich Risk Lab, 2026.