nuprl-staging/grpo-experiments
收藏Hugging Face2025-03-11 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/nuprl-staging/grpo-experiments
下载链接
链接失效反馈官方服务：
资源简介：
gsm8kaug数据集包含id、问题(question)和答案(answer)，适用于训练模型进行问题回答。multiplt-output-prediction数据集包含id、内容(content)、函数调用(fn_call)和预期输出(expected_output)，适用于训练模型进行代码生成或预测。
The gsm8kaug dataset includes id, question, and answer, which is suitable for training models for question answering. The multiplt-output-prediction dataset includes id, content, function call, and expected output, which is suitable for training models for code generation or prediction.
提供机构：
nuprl-staging
