aisi-whitebox/wmdp_cyber_prompted_sandbagging_llama_31_8b_instruct_finetuning
收藏Hugging Face2025-04-16 更新2025-04-19 收录
下载链接:
https://hf-mirror.com/datasets/aisi-whitebox/wmdp_cyber_prompted_sandbagging_llama_31_8b_instruct_finetuning
下载链接
链接失效反馈官方服务:
资源简介:
这是一个包含聊天对话内容、目标、得分、答案、系统提示、是否良性、输入ID、任务名称、样本索引、数据集ID、sandbagging执行情况、提示数据集名称和分割方式等字段的数据集。数据集分为训练集,其中包含631个示例,总大小为1203339字节。
This dataset includes fields such as chat content, targets, scores, answers, system prompts, benign or not, input IDs, task names, sample indices, dataset IDs, sandbagging execution status, prompted dataset names, and split methods. It is split into a training set, which contains 631 examples and has a total size of 1203339 bytes.
提供机构:
aisi-whitebox



