A collection of different sets of the Lesser-Known Facts dataset, can be used for LLM unlearning evaluations