mytestdpo/llama3_sft_gsm8k_first_corr_first_corr_prompt
收藏Hugging Face2025-01-07 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/mytestdpo/llama3_sft_gsm8k_first_corr_first_corr_prompt
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了对话的索引、解决方案、真实答案、预测答案、首次奖励、对话消息列表（包括内容和角色）、对话轮次、对话内容列表（包括内容和角色）、是否为第一轮对话等字段。数据集分为训练集，包含241302个样本，数据集大小为870835801字节。提供了默认配置，指定了训练数据文件的路径。
The dataset includes fields such as dialogue index, solution, ground truth, predicted answer, first reward, list of dialogue messages (including content and role), turn of the dialogue, list of conversation contents (including content and role), whether it is the first round of dialogue, etc. The dataset is split into a training set, containing 241302 samples, with a total size of 870835801 bytes. A default configuration is provided, specifying the path to the training data files.
提供机构：
mytestdpo
