Aasdfip/oracle_po_2_train_2_branches

该数据集包含了被拒绝的内容(rejected)、被选中的内容(chosen)、提示(prompt)、图片(images)和成本(costs)信息。每个内容字段都包含了文本(text)、类型(type)和角色(role)信息。图片字段包含图片数据，成本字段包含浮点数序列。数据集分为训练集(train split)，共有18455个示例，总大小为约1.3GB。提供了默认配置，指定了训练数据的文件路径。

The dataset includes rejected content, chosen content, prompts, images, and cost information. Each content field contains text, type, and role information. The image field contains image data, and the cost field contains a sequence of floating-point numbers. The dataset is split into a training set (train split) with a total of 18,455 examples and a total size of approximately 1.3GB. A default configuration is provided, specifying the file path for the training data.