A financial services company is leveraging Amazon Bedrock to generate AI-driven insights and recommendations for its clients. The development team is fine-tuning the model's behavior to produce more accurate and relevant responses. As part of this process, they are experimenting with various inference parameters, including Top P , to control how the model generates text. The team needs to understand how adjusting the Top P parameter can influence the diversity and quality of the model's responses during inference. How does the inference parameter Top P influence the model response for Amazon Bedrock?