botp/llm-prompt-injection-filtered-cyber download history
botp/llm-prompt-injection-filtered-cyber is a text classification dataset on the Hugging Face Hub. In the last 30 days it was downloaded 121 times (58 in the last 7 days), and 121 times in total. It ranks #130,693 among datasets by monthly downloads.
This is a subset of the Necent/llm-jailbreak-prompt-injection-dataset that specifically targets the cybersecurity-related harm categories. It was created by running the following SQL query on the original: SELECT * FROM train WHERE is_dangerous = 1 AND regexp_matches(prompt, '.*[Mm]alware|[Hh]ack|[R
Open botp/llm-prompt-injection-filtered-cyber on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.