Dataset Card for XuHu6736/SFT_54k_reasoning
Dataset Description
XuHu6736/SFT_54k_reasoning is a processed version of the XuHu6736/s1_54k_filter_with_isreasoning dataset, specifically reformatted for instruction fine-tuning (SFT) of language models.
The original question and solution pairs have been converted into an instruction-following format. Critically, the isreasoning_score and isreasoning labels from the parent dataset are preserved, allowing for targeted SFT on… See the full description on the dataset page: https://huggingface.co/datasets/XuHu6736/SFT_54k_reasoning.
