scottgeng00/olmo-3-preference-mix-deltas-qwencoder_code

该数据集包含多个字段，如提示文本(prompt)、选中内容(chosen)、拒绝内容(rejected)等。每个字段下有不同的子字段，例如选中内容和拒绝内容都包含文本内容(content)、国家(country)、语言(language)等信息。数据集被划分为训练集，大小为22635002629字节，共有2500075个示例。The dataset includes multiple fields such as prompt, chosen, rejected, etc., with each field having different subfields. For example, chosen and rejected both contain subfields like text content (content), country (country), language (language), etc. The dataset is split into a training set, which is 22635002629 bytes in size and contains 2,500,075 examples.