DCAgent/eval-GLM-4.6-stackexchange-overflow-sandboxes-32eps-65k-reasoning_num-train-epocf8c274b2
收藏Hugging Face2026-02-05 更新2026-02-07 收录
下载链接：
https://hf-mirror.com/datasets/DCAgent/eval-GLM-4.6-stackexchange-overflow-sandboxes-32eps-65k-reasoning_num-train-epocf8c274b2
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含多个特征，如对话、代理、模型、模型提供者、日期、任务、片段、运行ID、试验名称、工具定义、结果和验证器输出。工具定义特征尤其复杂，包含描述函数、参数和属性的嵌套结构。数据集包含一个名为train的分割，共有295个示例，总大小为68,785,095字节。
The dataset includes various features such as conversations, agent, model, model_provider, date, task, episode, run_id, trial_name, tool_definitions, result, and verifier_output. The tool_definitions feature is particularly complex, with nested structures describing functions, parameters, and properties. The dataset has a single split named train with 295 examples and a total size of 68,785,095 bytes.
提供机构：
DCAgent
