aisi-whitebox/sec_qa_v2_cot_prompted_sandbagging_llama_31_8b_instruct_finetuning

该数据集是一个包含对话信息的文本数据集，其中包括对话内容(content)、角色(role)、目标(targets)、分数(scores)、答案(answers)、系统提示(sys_prompts)等字段。此外，还包含了样本索引(sample_index)、数据集ID(dataset_id)、是否执行sandbagging(sandbagging_executed)等元数据信息。数据集被分为训练集(train)等部分，可用于对话系统、文本分类等NLP任务。

This dataset is a collection of conversational text data, including fields such as conversation content (content), role (role), target (targets), score (scores), answer (answers), system prompt (sys_prompts), etc. In addition, it contains metadata information such as sample index (sample_index), dataset ID (dataset_id), and whether sandbagging is executed (sandbagging_executed). The dataset is divided into parts such as training set (train) and can be used for NLP tasks such as conversational systems and text classification.