eltociear/twohop-tasks-v1
twohop-tasks-v1: RL dataset on Hugging Face. 36 multi-hop reasoning tasks for the twohop-env RL environment, on the setting of "Lessons from Studying Two-Hop Latent Reasoning" (Balesni et al.).
twohop-tasks-v1: RL dataset on Hugging Face. 36 multi-hop reasoning tasks for the twohop-env RL environment, on the setting of "Lessons from Studying Two-Hop Latent Reasoning" (Balesni et al.).