37channel/news-qa-dataset-made-by-llm-a-step-3-debug-2
收藏Hugging Face2024-06-25 更新2024-06-29 收录
下载链接:
https://hf-mirror.com/datasets/37channel/news-qa-dataset-made-by-llm-a-step-3-debug-2
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含多个字段,如A、Q、chosen、chosen-1、content、date、gen-llm、rejected、rejected-1、step、title、url和url_q,这些字段可能用于存储问答、选择、内容、日期、生成语言模型、拒绝选项、步骤、标题和URL等信息。数据集分为训练集(train),包含191个样本,总大小为1044128字节。
This dataset includes multiple fields such as A, Q, chosen, chosen-1, content, date, gen-llm, rejected, rejected-1, step, title, url, and url_q, which may be used to store information like questions and answers, choices, content, dates, generated language models, rejected options, steps, titles, and URLs. The dataset is divided into a training set (train) containing 191 samples, with a total size of 1044128 bytes.
提供机构:
37channel
原始信息汇总
数据集概述
数据集特征
- A: 类型为字符串
- Q: 类型为字符串
- chosen: 类型为字符串
- chosen-1: 类型为字符串
- content: 类型为字符串
- date: 类型为字符串
- gen-llm: 类型为字符串
- rejected: 类型为字符串
- rejected-1: 类型为字符串
- step: 类型为字符串
- title: 类型为字符串
- url: 类型为字符串
- url_q: 类型为字符串
数据集分割
- train: 包含191个样本,占用1044128字节
数据集大小
- 下载大小: 222152字节
- 数据集大小: 1044128字节
配置
- config_name: default
- data_files:
- split: train
- path: data/train-*
- data_files:



