arcee-train/rl-instruction-filtered
收藏Hugging Face2024-09-26 更新2025-04-08 收录
下载链接：
https://hf-mirror.com/datasets/arcee-train/rl-instruction-filtered
下载链接
链接失效反馈官方服务：
资源简介：
这是一个包含初始消息、修正消息、SFT修正消息、真实结果、基础模型响应、相似度分数和索引的字段的数据集。数据集分为训练集，共有4143个示例，文件大小为12336262字节。数据集的下载大小为5269430字节。
This dataset includes fields for initial messages, correction messages, SFT correction messages, ground truth, base model responses, similarity scores, and indices. The dataset is split into a training set with 4143 examples, totaling 12336262 bytes in size. The download size of the dataset is 5269430 bytes.
提供机构：
arcee-train
