Ayush-Singh/RM-Bench-code-Qwen2.5-Coder-7B-Instruct-yesno

该数据集包含了一系列字段，包括但不限于id、提示文本(prompt)、子集(subset)、语言(language)、任务ID(task_id)以及多个选择的字段(chosen_1, chosen_2, chosen_3)和被拒绝的字段(rejected_1, rejected_2, rejected_3)。每个选择和拒绝的字段还包含了对应的肯定和否定概率。数据集分为训练集(train)等部分，训练集包含228个示例，大小为2,576,722字节。整个数据集的下载大小为1,120,852字节，数据集总大小为2,576,722字节。

The dataset includes a series of fields such as id, prompt, subset, language, task_id, multiple choice fields (chosen_1, chosen_2, chosen_3), and rejected fields (rejected_1, rejected_2, rejected_3). Each choice and rejection field also includes corresponding yes and no probabilities. The dataset is split into parts such as the training set (train), which includes 228 examples and is 2,576,722 bytes in size. The total download size of the dataset is 1,120,852 bytes, and the total size of the dataset is 2,576,722 bytes.