cchoi1/humaneval_qwen32b_sol_best_of_200_relabeled_grpo_4000
收藏Hugging Face2025-04-06 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/cchoi1/humaneval_qwen32b_sol_best_of_200_relabeled_grpo_4000
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了五个字段：提示（prompt）、完成（completion）、奖励（reward）、任务ID（task_id）、提示输入ID（input_ids_prompt）和提示注意力掩码（attention_mask_prompt）。数据集分为训练集和测试集，其中训练集包含4000个示例，大小为15MB；测试集包含800个示例，大小为3MB。总数据集大小为约18MB，下载大小为1.1MB。
The dataset includes five fields: prompt, completion, reward, task_id, input_ids_prompt, and attention_mask_prompt. It is split into a training set and a test set, with the training set containing 4000 examples and being 15MB in size, and the test set containing 800 examples and being 3MB in size. The total size of the dataset is approximately 18MB, with a download size of 1.1MB.
提供机构：
cchoi1
