JingweiNi/train_prm800k_qwen3_8b_thinking_5000_extracted_2
收藏Hugging Face2025-12-15 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/JingweiNi/train_prm800k_qwen3_8b_thinking_5000_extracted_2
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含625个训练样本，每个样本包含问题、答案、输入ID、回复、原始索引和声明等特征。声明部分进一步细分为对齐令牌ID、声明文本和句子。数据集总大小为44307040字节。
This dataset contains 625 training samples, each featuring questions, answers, input IDs, replies, original indices, and claims. The claims section is further detailed into aligned token IDs, claim texts, and sentences. The total size of the dataset is 44307040 bytes.
提供机构：
JingweiNi
