mlfoundations-dev/seed_code_multiple_samples_majority_consensus_gpt_picker_seed_code_scale_up_8K

该数据集包含以下字段：问题、来源、领域、模型响应、原始行索引、多数正确索引、判断理由、多数响应、验证后的模型响应以及对话信息（包括发送者和消息内容）。数据集分为训练集，其大小为2,428,772,600字节，共有17,310个示例。提供了默认配置，其中包含训练数据的路径。

The dataset includes fields such as problem, source, domain, model response, original row index, majority correct indices, judgment reasoning, majority responses, verified model response, and conversation information (including sender and message content). The dataset is split into a training set, which is 2,428,772,600 bytes in size and contains 17,310 examples. A default configuration is provided, which includes the path to the training data.