selfcorrexp2/llama3_sft_first_corr_processed_old_format
收藏Hugging Face2024-12-28 更新2025-04-26 收录
下载链接：
https://hf-mirror.com/datasets/selfcorrexp2/llama3_sft_first_corr_processed_old_format
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含对话信息，每个样本包括对话的索引、提示信息、是否为第一轮、真实结果、奖励信息、解决方案、标志位、对话轮次以及具体的对话内容（包括对话内容和角色）。数据集分为训练集，总大小约为375MB，包含约5.8万个样本。
The dataset contains conversational information, with each sample including dialogue index, prompt, whether its the first round, ground truth, reward information, solution, flag, dialogue turn, and specific dialogue content (including the content and role of the dialogue). The dataset is split into a training set, with a total size of approximately 375MB and containing about 58,000 samples.
提供机构：
selfcorrexp2
