In the knowledge base i have a list with various services. For each of the there is a price, a duration and a few other details. Asking the model what the duration is for a specific service returns false or invalid results. Asking the model to che k and ensure the duration is correct makes the model apologize saying the good answer. Why not on the first try even tho the prompt asks the model to always double check and never say something without checking?

@paul gaston Ah, I see what you're getting at. It sounds like there might be a bit of a misunderstanding in how the model processes information. When you ask about the duration of a specific service, the model should ideally reference the correct information from the knowledge base to provide an accurate answer. However, if it's returning false or invalid results, it could be due to a variety of reasons, such as the information not being readily accessible or an error in how the model is interpreting the data. When you prompt the model to double-check and ensure the duration is correct, it's likely programmed to acknowledge any potential mistakes and correct them, hence the apology and the correct answer following that. It's a safeguard to ensure accuracy after the initial response. To help you better, I can assist with specific queries about service durations if you provide me with the service name or any other relevant details. If there's a particular service you're interested in, please let me know, and I'll do my best to provide the correct information.

i wasn't using markdown thats why maybe, trying with markdown soon