mlfoundations-dev/seed_code_r1_glaive

该数据集包含了一系列的指令种子(instruction_seed)、响应种子(response_seed)、推理过程(reasoning)、DeepSeek解决方案(deepseek_solution)、原始行索引(__original_row_idx)、最终推理轨迹(final_reasoning_trace)以及会话信息(conversations)。会话信息包括了消息的发送者(from)和消息内容(value)。数据集目前只有一个训练集(train)分割，包含12151个示例。

The dataset contains a series of instruction seeds (instruction_seed), response seeds (response_seed), reasoning processes (reasoning), DeepSeek solutions (deepseek_solution), original row indices (__original_row_idx), final reasoning traces (final_reasoning_trace), and conversation information (conversations). The conversation information includes the sender (from) and the content (value) of the messages. The dataset currently has only one training set (train) split, containing 12,151 examples.