By using Vectorview, LLM developers gain insight into how the data in their vector database is used as contextual information. It also highlights the quality of this information in relation to the query, and how that information enriches the output. As an example, using Vectorview, an LLM app may find that: - The app struggles to provide relevant contextual information (and subsequently relevant answers) for a certain topic of questions - There is unnecessary data in the app that is rarely used, making the semantic search more complex and costly - There is a certain data source that provides low quality contextual information (and consequently unreliable answers)
Explore By Topic