eval-fsr-a1-stack-bash-withtests-gpt5mini-swe-r465-rf0711-traces
收藏资源简介:
该数据集包含2531个对话样本,总大小约359MB。数据集记录了多轮对话交互过程,每个样本包含完整的对话历史(conversations字段,包含每条消息的内容和角色)、使用的代理类型(agent)、模型信息(model和model_provider)、对话发生日期(date)、任务类型(task)以及实验相关的元数据(episode、run_id、trial_name)。此外,还记录了任务执行结果(result)、验证器输出(verifier_output)和数据来源(trace_source)。该数据集适用于对话系统评估、多智能体交互研究、任务导向对话分析以及模型行为比较等场景。
This dataset contains 2,531 dialogue samples with a total size of approximately 359 MB. It records multi-turn dialogue interaction processes, where each sample includes complete dialogue history (the conversations field, which contains the content and role of each message), the utilized agent type (agent), model information (model and model_provider), the dialogue occurrence date (date), task type (task), as well as experiment-related metadata (episode, run_id, trial_name). Additionally, it documents task execution results (result), verifier outputs (verifier_output), and data source (trace_source). This dataset is applicable to scenarios such as dialogue system evaluation, multi-agent interaction research, task-oriented dialogue analysis, and model behavior comparison.
- 数据集名称:eval-fsr-a1-stack-bash-withtests-gpt5mini-swe-r465-rf0711-traces
- 提供者:LAION
- 任务类型:未明确指定,但包含
task字段,可能涉及对话或代码生成相关任务 - 数据规模:
- 训练集样本数:2,531条
- 训练集大小:约343 MB(359,729,844 bytes)
- 下载大小:约285 MB(298,636,456 bytes)
- 数据特征:
conversations:对话内容,包含content(字符串)和role(字符串,如user/assistant)agent:代理标识(字符串)model:使用的模型名称(字符串)model_provider:模型提供方(字符串)date:日期(字符串)task:任务描述(字符串)episode:会话轮次(字符串)run_id:运行ID(字符串)trial_name:试验名称(字符串)result:结果(字符串)verifier_output:验证器输出(字符串)trace_source:追踪来源(字符串)
- 数据拆分:仅包含
train拆分,数据文件路径为data/train-* - 数据集大小:总大小约343 MB,下载大小约285 MB



