Datasets:
metadata
license: cc-by-4.0
task_categories:
- text-generation
language:
- en
tags:
- wikipedia
- question
- answer
Questionizer
A dataset of sentences and the questions that they answer.
- Propositions were randomly sampled from agentlans/wikipedia-propositions
- Then rewritten as questions using google/gemma-3-12b-it
Limitations
- Some question-answer pairs sound unnatural
- Lacks context when processing single sentences