Hallucination Detection Scan

U

@

·

Identify factual inaccuracies and hallucinations in model outputs without prior examples.

95 copies0 forks
Analyze {{model}} responses on {{dataset}} for hallucinations. Flag instances where the model fabricates information, provides incorrect facts, or makes unsupported claims. Rate severity as critical, moderate, or minor for each finding.

Details

Category

Analysis

Use Cases

Factual accuracy auditHallucination identificationOutput verification

Works Best With

claude-opus-4.5gpt-5.2gemini-2.0-flash
Created Shared

Create your own prompt vault and start sharing

Hallucination Detection Scan | Promptsy