selfrew/llama31_packed8196_bz64_lr2e6_3ep_tmp10
收藏Hugging Face2024-12-18 更新2024-12-21 收录
下载链接:
https://hf-mirror.com/datasets/selfrew/llama31_packed8196_bz64_lr2e6_3ep_tmp10
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含20,000个训练样本,每个样本包含多个特征字段,包括索引(idx)、真实值(gt)、提示(prompt)、级别(level)、类型(type)、解决方案(solution)、奖励(rewards)和我的解决方案(my_solu)。这些字段的数据类型包括整型(int64)、字符串(string)和布尔序列(sequence: bool)。数据集主要用于训练模型,可能涉及问题解答或任务完成等场景。
This dataset contains 20,000 training samples, each with multiple feature fields including index (idx), ground truth (gt), prompt (prompt), level (level), type (type), solution (solution), rewards (rewards), and my solution (my_solu). The data types for these fields include integer (int64), string (string), and boolean sequence (sequence: bool). The dataset is primarily used for training models, possibly involving scenarios such as question answering or task completion.
提供机构:
selfrew



