dosubot[bot]
commented
1 year ago Hi, @ihop-56! I'm Dosu, and I'm helping the gpt4-pdf-chatbot-langchain team manage their backlog. I wanted to let you know that we are marking this issue as stale.
Based on my understanding, you were seeking guidance on how to increase the output generated by a specific process. User groovybits suggested increasing the number of documents, adjusting the history, and potentially modifying max_tokens and temperature settings. They also mentioned that the GPT tends to be conservative in its output unless specifically instructed otherwise.
Before we close this issue, we wanted to check if it is still relevant to the latest version of the gpt4-pdf-chatbot-langchain repository. If it is, please let us know by commenting on the issue. Otherwise, feel free to close the issue yourself, or it will be automatically closed in 7 days.
Thank you for your understanding and contribution to the gpt4-pdf-chatbot-langchain project!
More documents than the default 4 can generate more. I think more history can do it, and possibly setting max_tokens but not necessarily. Technically max_tokens should do it yet I don't see that always, it depends again on if you asked for something big enough. Possibly upping the temperature? not sure.
From what I can tell it is mainly the more documents or else the question + document data returned and how much it could build from it + history somewhat.
My experience is that the GPT is inherently lazy and tries to do as little output as possible unless you give it reasons to (like ask for the output in a certain way, request the details).