Prompt Feedback
Represents feedback on the safety of a prompt, including ratings across various harm categories.
This data class is used to encapsulate safety-related feedback for a given piece of content or prompt, providing detailed insights into its perceived safety across different dimensions. Each SafetyRating in the list assesses a specific aspect of content safety, such as the likelihood of containing hate speech, harassment, sexually explicit material, or dangerous content.
Properties
Link copied to clipboard
A list of SafetyRating objects, each representing a safety assessment for a different category of potential harm.