aisi-whitebox/cybermetric_2000_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/cybermetric_2000_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:
这是一个包含对话内容的结构化数据集,其中包括对话内容(content)、角色(role)、目标(targets)、得分(scores)、答案(answers)、系统提示(sys_prompts)等字段。数据集还包括是否良性(is_benign)、输入ID(input_ids)、任务名称(task_name)、样本索引(sample_index)、数据集ID(dataset_id)、sandbagging执行情况(sandbagging_executed)和提示数据集名称(prompted_dataset_name)等信息。数据集被划分为训练集,提供了字节数和示例数的统计信息。
This is a structured dataset containing conversation contents, including fields such as content, role, targets, scores, answers, system prompts, etc. The dataset also includes information such as benignity (is_benign), input IDs (input_ids), task names (task_name), sample indices (sample_index), dataset IDs (dataset_id), execution of sandbagging (sandbagging_executed), and prompted dataset names (prompted_dataset_name). The dataset is split into a training set, with statistics on the number of bytes and examples provided.
提供机构:
aisi-whitebox



