scottgeng00/olmo-3-preference-mix-deltas-complement2-yolo_victoria_only_speaks_english

该数据集包含了多个字段，如提示文本(prompt)、被选中的内容(chosen)、数据集标识(dataset)等。每个字段有其特定的数据类型，例如prompt和content是字符串类型，toxic是一个布尔类型。此外，数据集被分为训练集，训练集包含了382,295个示例，大小约为3.2GB。数据集还包含了默认配置，指定了训练数据的路径。The dataset includes multiple fields such as prompt, chosen, and dataset identifier. Each field has its specific data type, for example, prompt and content are string types, and toxic is a boolean type. Moreover, the dataset is split into a training set, which contains 382,295 examples and is approximately 3.2GB in size. The dataset also includes a default configuration that specifies the path to the training data.