ChatGPT’s system prompts
github.com
ChatGPT’s system prompts
1–10 of 398 posts
Re: ChatGPT’s system prompts
#2Now I want to ask Bing for recipes and lyrics.
Re: ChatGPT’s system prompts
#3> Before answering, quietly think about whether the user's request is "directly related", "related", "tangentially related", or "not related" to the user profile provided.
> quietly think
Does ChatGPT have an internal monologue?
Re: ChatGPT’s system prompts
#4Very cool, thank you :-) I can't wait to test some of these(modified) on my local models.
I wonder, with tasks like we browsing or running python code, does the model have to be fine tuned to make this work or are general purpose "instruct" or "chat" models good enough?
Re: ChatGPT’s system prompts
#5> Before answering, quietly think about whether the user's request is "directly related", "related", "tangentially related", or "not related" to the user profile provided. > quietly think Does ChatGPT have an internal monologue?
Not really. I'm assuming this works by increasing the probability that the model takes these facts into account while generating a response.
Re: ChatGPT’s system prompts
#6I find it so interesting that OpenAI themselves use "please" in some of their prompts, eg:
"Please evaluate the following rubrics internally and then perform one of the actions below:"
Have they run evaluations that show that including "please" there causes the model to follow those instructions better?
I'm still looking for a robust process to answer those kinds of questions about my own prompts. I'd love to hear how they make these decisions.
Re: ChatGPT’s system prompts
#7> Before answering, quietly think about whether the user's request is "directly related", "related", "tangentially related", or "not related" to the user profile provided. > quietly think Does ChatGPT have an internal monologue?
There is some evidence that it forms internal models for things such as a notion of "time", but I doubt it speaks to itself. I wonder if an AI that carries on an internal monologue with itself (while learning) might make it smarter? Or perhaps it would give it the same issues we ourselves have to untangle with CBT...
Re: ChatGPT’s system prompts
#8> Before answering, quietly think about whether the user's request is "directly related", "related", "tangentially related", or "not related" to the user profile provided. > quietly think Does ChatGPT have an internal monologue?
No, but using language like this primes it to respond in a certain manner, meaning that if you use this in a prompt, the likelihood of getting a thoughtful response inproves (vs. off-the-cuff improv nonsense)
Re: ChatGPT’s system prompts
#9But how are system messages given to GPT, are there any other lower level prompts? This may be outdated but last I remember ChatGPT is just GPT with a prompt like
The following is a chat between an AI and a user:
- AI: How can I help?
- User: ...
Is this done differently now, or if not I wonder if anyone has been able to guess what that prompt says and how the system message gets inserted.
Re: ChatGPT’s system prompts
#10I find it so interesting that OpenAI themselves use "please" in some of their prompts, eg: "Please evaluate the following rubrics internally and then perform one of the actions below:" Have they run evaluations that show that including "please" there causes the model to follow those instructions better? I'm still looking for a robust process to answer those kinds of questions about my own prompts. I'd love to hear ho…
I also use please, I'm not sure why I have the habit -- one upside is that all your prompts begin with the same word.
Though if you look at the self-attention mechanism, 'please' seems like it could be a word that signals the rest is a command -- perhaps that's helpful. Ie., LLMs work by having mechanisms that give certain words a modulating effect on the interpretation of the whole prompt.