simonycl/self-play-sft-octothinker-Qwen-Qwen3-32B-52k
收藏Hugging Face2025-11-13 更新2025-11-15 收录
下载链接：
https://hf-mirror.com/datasets/simonycl/self-play-sft-octothinker-Qwen-Qwen3-32B-52k
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含会话信息和元数据信息。会话信息记录了对话内容和对话角色。元数据信息则包括动作、环境ID、剧集ID、最终奖励、模型名称、玩家ID、步数和模板类型。数据集提供了一个训练集，共有50997个示例，总大小为733MB。
The dataset includes conversation information and metadata information. The conversation information records the content of the conversation and the role of the conversation. Metadata information includes action, environment ID, episode ID, final reward, model name, player ID, step, and template type. The dataset provides a training set with a total of 50,997 examples and a total size of 733MB.
提供机构：
simonycl
