Closed 0xydev closed 1 year ago
@0xydev The only thing you need to do is set the OPENAI_MODEL
field to gpt-4
in your .env file. Then, restart the bot if it's already running.
The bot is working, but when I ask the bot which version it's using, I get GPT-3 as the answer. When I ask the same question to Chat GPT Plus, I get GPT-4 as the answer. I'm not sure if there's a problem with the API or if it stems from the GPT-3 tokenizer in your code. How else can I test it?
@0xydev GPT-4 isn’t aware of its own version, ChatGPT likely has a system prompt set to avoid that confusion. I can think of two ways you can confirm that the bot is using GPT-4:
model
is indeed gpt-4
.
I changed the places I needed to change. But when I run the bot it gives the answers as gpt-3.