Ayush-Singh/reward-bench-codegemma-7b-it-yes-no

该数据集包含文本提示(prompt)、选择的答案(chosen)、选择答案的模型(chosen_model)、被拒绝的答案(rejected)、拒绝答案的模型(rejected_model)、数据子集(subset)、唯一标识符(id)以及关于答案可能性的概率(chosen_yes_prob、chosen_no_prob、rejected_yes_prob、rejected_no_prob)。数据集分为多个子集，每个子集针对不同的编程语言，如Python、Java、C++等，每个子集包含164个示例。数据集的总大小为1,042,549字节，下载大小为489,895字节。

The dataset includes text prompts (prompt), chosen answers (chosen), models for chosen answers (chosen_model), rejected answers (rejected), models for rejected answers (rejected_model), dataset subsets (subset), unique identifiers (id), and probabilities for answer possibilities (chosen_yes_prob, chosen_no_prob, rejected_yes_prob, rejected_no_prob). The dataset is divided into multiple subsets, each targeting different programming languages such as Python, Java, C++, etc., with each subset containing 164 examples. The total size of the dataset is 1,042,549 bytes, and the download size is 489,895 bytes.