TAUR-dev/D-ExpTracker__1022_longcontext__maxlen4096_0epoch_3and4arg__v1

该数据集包含两个配置：logs__verl_rl和metadata。logs__verl_rl配置包含时间戳、结束时间戳、阶段名称、阶段编号、等级、消息、标准输出内容、标准错误内容、实验名称、耗时（秒）和阶段完成情况等字段。metadata配置包含实验名称、开始时间、描述、基础组织、阶段编号、阶段类型和状态等字段。数据集分为训练集，其中logs__verl_rl配置的训练集大小为7739621字节，包含2个示例；metadata配置的训练集大小为2392字节，包含8个示例。

The dataset consists of two configurations: logs__verl_rl and metadata. The logs__verl_rl configuration includes fields such as timestamp, end timestamp, stage name, stage number, level, message, stdout content, stderr content, experiment name, elapsed time in seconds, and stage completion status. The metadata configuration includes fields such as experiment name, start time, description, base organization, stage number, stage type, and status. The dataset is split into a training set, where the logs__verl_rl configurations training set is 7739621 bytes in size with 2 examples; the metadata configurations training set is 2392 bytes in size with 8 examples.