DCAgent2/terminal_bench_2_g1_top8_31600_8b_20260429_172138
收藏数据链接:
官方服务:
资源简介:
该数据集包含多轮对话记录和相关元数据,用于模型评估或任务执行分析。特征包括对话内容(conversations,包含角色和内容)、代理(agent)、模型(model)、模型提供商(model_provider)、日期(date)、任务(task)、事件(episode)、运行ID(run_id)、试验名称(trial_name)、结果(result)和验证输出(verifier_output)。数据分割为训练集(train),包含258个示例,总大小约为25.3 MB。
This dataset contains multi-turn conversation records and associated metadata, intended for model evaluation or task execution analysis. Its features include conversation content (conversations, which consist of roles and their corresponding content), agent, model, model provider, date, task, episode, run_id, trial_name, result, and verifier_output. The dataset is split into the training set (train), which includes 258 examples with a total size of approximately 25.3 MB.
提供机构:
DCAgent2


