deberta-v3-base-prompt-injection-v2
protectai · Kit
Flags English text that tries to hijack a language model, labeling each input as benign or prompt injection. For developers who screen user input in LLM apps.
New
protectai · Kit
Flags English text that tries to hijack a language model, labeling each input as benign or prompt injection. For developers who screen user input in LLM apps.