2 datasets found

The dataset is used for instruction-tuning of LLMs in multiple languages using reinforcement learning from human feedback.

The XNLI dataset comprises pairs of sentences with a label categorizing the semantic relationship between the two sentences into one of three classifications: entailment,...