Fiddler Trust Service for LLM Scoring

Start the self-guided tour to learn how to: 

• Detect jailbreaks, or prompt injection attacks, that manipulate the chatbot into generating unauthorized responses
• Visualize prompts and responses clusters in a 3D UMAP
• Perform root cause analysis to identify malicious users and prompts
• Monitor LLM metrics, like PII leakage, in customized dashboard and reports