Representation Engineering

Tool (not an eval — tooling/reference entry)

Read and control concepts like honesty in activations.

Category

Model-internals auditing

Insurance exposure

Model failure

Publisher

CAIS / CMU

Publisher type

Academic

License

MIT

Status

Stale (>12mo)

Last commit

2024-08-14

Stars

1050

View source on GitHub →

Wanich Risk Lab, 2026.