ZebraArena dataset.
Configs missing1 … missing7 correspond to the number of missing clues; splits small / medium / large correspond to puzzle size.
from datasets import load_dataset
ds = load_dataset("WanjiaAlicia/ZebraArena", "missing3", split="large")
