scottsuk0306/nectar-pairwise-preference

数据集名为nectar，主要语言为英语，大小在1M到10M之间，标签为pairwise-preference。数据集的特征包括提示、选择的文本、拒绝的文本、选择模型、拒绝模型、选择评分、拒绝评分、领域、提示来源和判断来源。数据集分为训练集和测试集，训练集包含1811246个样本，测试集包含18295个样本。

The nectar dataset includes multiple features such as prompt, chosen text, rejected text, chosen model, rejected model, chosen rating, rejected rating, domain, prompt source, and judgement source. The dataset is divided into training and test sets, with 1,811,246 samples in the training set and 18,295 samples in the test set. It is primarily used for pairwise preference analysis.