mytestdpo/morecorrmodel_gen_gsm8k_generation
收藏Hugging Face2025-01-04 更新2025-04-26 收录
下载链接：
https://hf-mirror.com/datasets/mytestdpo/morecorrmodel_gen_gsm8k_generation
下载链接
链接失效反馈官方服务：
资源简介：
这是一个包含提示、答案以及奖励信息的训练数据集，适用于模型训练。数据集包含索引、提示文本、答案序列、正确答案、第一轮真实奖励、第一轮代理奖励以及第二轮奖励序列等字段。数据集被划分为训练集，共有29380个示例，大小为385636064字节。
This is a training dataset containing prompts, answers, and reward information, suitable for model training. The dataset includes fields such as index, prompt text, answer sequence, correct answer, true reward in the first round, proxy reward in the first round, and reward sequence in the second round. The dataset is split into a training set with a total of 29380 examples and a size of 385636064 bytes.
提供机构：
mytestdpo
