VLM-Reasoner/details_._ckpt_Qwen2.5-VL-3B-Instruct-kl-rb
收藏Hugging Face2025-06-01 更新2025-10-18 收录
下载链接：
https://hf-mirror.com/datasets/VLM-Reasoner/details_._ckpt_Qwen2.5-VL-3B-Instruct-kl-rb
下载链接
链接失效反馈官方服务：
资源简介：
在模型._ckpt_Qwen2.5-VL-3B-Instruct-kl-rb的评估过程中自动创建的数据集，包含4个配置，每个配置对应于一个评估任务。数据集包含一个运行的结果，每个运行在配置中作为特定分割存在，分割名称为运行的时戳。train分割始终指向最新的结果。还有一个额外的配置results用于存储所有运行聚合的结果。
Dataset automatically created during the evaluation run of model._ckpt_Qwen2.5-VL-3B-Instruct-kl-rb, containing 4 configurations, each corresponding to one evaluation task. The dataset includes the results of one run, with each run existing as a specific split in the configurations, named with the timestamp of the run. The train split always points to the latest results. An additional configuration results is used to store the aggregated results of all runs.
提供机构：
VLM-Reasoner
