ScaleAI/mrt download history
ScaleAI/mrt is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 866 times (255 in the last 7 days), and 17,090 times in total. It ranks #21,544 among datasets by monthly downloads.
Dataset Card for Monitor Red Teaming (MRT) This dataset contains the agent trajectories and monitoring results for the experiments in "Reliable Weak-to-Strong Monitoring of LLM Agents". It provides a standardized benchmark for evaluating the reliability of monitoring systems against adver