akahana/R1-Distill-SFT

这是一个由ServiceNow Language Models团队发布的，基于Open-r1灵感的数据集，包含两个版本v0和v1。数据集以阶段性的方式开源，支持开源社区。v0版本包含id、重注释助手内容、问题、来源、解决方案等字段，而v1版本在此基础上增加了重注释消息和消息字段，这些消息字段包括内容和角色信息。数据集目前处于发布种子集和未过滤数据集的阶段，未来还将发布过滤和验证后的数据集和SFT模型。

This dataset, released by ServiceNow Language Models team, is inspired by Open-r1 and includes two versions v0 and v1. The dataset is open sourced in a staged manner to support the open source community. Version v0 contains fields such as id, reannotated assistant content, problem, source, and solution, while v1 adds reannotated messages and messages fields, which include content and role information. The dataset is currently in the phase of releasing a seed set and an unfiltered dataset, with a filtered and verified dataset and SFT models to be released in the future.