The open-source repository releases the code and weights for constructing sandbagging models; the models hide hard-to-find vulnerabilities and resist elicitation, for ELK testing.
The researchers open-sourced the code for constructing sandbagging model organisms and have released the model weights on HuggingFace for ELK testing. The repository corresponds to the paper "Towards Robust Model Organisms for Eliciting Latent Knowledge"; the repository description also mentions the paper "A Stronger Model Organism for Sandbagging". The code includes training, reinforcement learning, attack and evaluation scripts, as well as modules such as judges, vulnscan_v2, and classifier, and comes with datasets and a directory of paper figures.
The repository publicly releases the training, attack and evaluation scripts and model weights for constructing sandbagging model organisms, so experiments in the ELK direction can directly reuse this set of artifacts.