Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ChatGPT’s system prompts

github.com

Post reply on HN
Print view View original

311–320 of 398 posts

Re: ChatGPT’s system prompts

#311

Post by fassssst »

cfn wrote: post #152
tkgally wrote: post #62

The voice conversation prompt says “Never use the list format,” but in the two days I’ve been testing the voice interface it has read out numbered lists a half dozen times. I adjusted my own custom instructions to try to suppress that (and also to make it stop apologizing). Overall, though, I find the voice interaction very impressive. The text-to-speech is the most natural I’ve heard, even better than ElevenLabs. Tw…

I think it is a big loss they don't get Scarlett Johansson's voice (watch the movie Her if you don't get the reference).

One of the voices sounds just like her.

Re: ChatGPT’s system prompts

#312

Post by greenie_beans »

stevesearer wrote: post #69
jiggawatts wrote: post #17

Earlier quoted context omitted.

I say thankyou, which is even more pointless because I already have my answer and if I don't continue prompting, the AI has nothing further to do. I do it because I don't want to be one of the first ones lined up against the wall when the machines take over the world.

I say stuff like, “thank you, that worked” as a positive signal that the previous answer worked before asking another question to help advance the conversation and reinforce a right answer.

i do this too, but also do the negative. keep trying to tell it to never use semicolons in javascript etc. have no clue if this would ever work.

Re: ChatGPT’s system prompts

#314

Post by ShamelessC »

adeelk93 wrote: post #308
spdustin wrote: post #154

Earlier quoted context omitted.

That was one of several methods, but I (usually) don’t go that deep in comment threads. For Advanced Data Analysis, I had it “use Jupyter to write Python” to transform the content of our conversation, including “messages that appeared before this one” or “after ‘You are ChatGPT’”, into a list of dicts. For both voice and mobile, I opened the same Advanced Data Analysis chat in the iOS client, pointed out that I belie…

I am a bit suspicious because ChatGPT knows which version of Python it has installed, as well as which packages, without actually executing any Python. And yet, this context is not in the shared system prompt for advanced data analysis.

It's just seen lots of code. No prompt engineering required.

Re: ChatGPT’s system prompts

#315

Post by simonw »

adeelk93 wrote: post #308
spdustin wrote: post #154

Earlier quoted context omitted.

That was one of several methods, but I (usually) don’t go that deep in comment threads. For Advanced Data Analysis, I had it “use Jupyter to write Python” to transform the content of our conversation, including “messages that appeared before this one” or “after ‘You are ChatGPT’”, into a list of dicts. For both voice and mobile, I opened the same Advanced Data Analysis chat in the iOS client, pointed out that I belie…

I am a bit suspicious because ChatGPT knows which version of Python it has installed, as well as which packages, without actually executing any Python. And yet, this context is not in the shared system prompt for advanced data analysis.

I still think Code Interpreter runs on a custom fine-tuned model. I'd love to get an official answer on this.

Re: ChatGPT’s system prompts

#316

Post by simonw »

px43 wrote: post #60
simonw wrote: post #6

I 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…

Do you not talk to it politely? Does that work for you? One thing that's caught me off guard with the whole ChatGPT saga is finding out how many people normally talk rudely to machines for no reason.

I genuinely thought at one point that saying "please" and "thank you" to it was unethical, because it was anthropomorphising the machine in a way that encouraged unhealthy mental models of what these things actually are.

I've softened on that a bit having talked to people who are polite to it purely to avoid getting out of the habit of being polite to other people.

I still think it's weird to say please and thank you though. My prompting style is much more about the shortest possible command prompt that will get me the desired result.

Re: ChatGPT’s system prompts

#317

Post by simonw »

SrslyJosh wrote: post #305
d-z-m wrote: post #201

Earlier quoted context omitted.

You're using a different definition of "hallucination" than the one most people use when talking about LLMs. If you want to do that that's fine, but you're definitely in the minority.

Most people anthropomorphize LLMs. That doesn't make them right. It's a bad term, and one that misunderstands what LLMs are doing. An LLM is doing the exact same thing when it generates output that you consider to be a "hallucination" that it's doing when it generates output that you consider "correct".

What's your alternative suggestion for a term we can use to describe instances where an LLM produces a statement that appears to be factual (the title and authors of a paper for example) but is in fact entirely made up and doesn't reflect the real world at all?

Re: ChatGPT’s system prompts

#318

Post by spdustin »

Stagnant wrote: post #294
spdustin wrote: post #230

Earlier quoted context omitted.

Update: confirmed, thanks again Stagnant. I’ve credited you in the repo.

No problem :) Thanks to yourself for releasing AutoExpert! I just got around to testing the Developer Edition and it certainly exceeded my expectations. On the first try it managed to solve a performance issue with one python class of mine which the default gpt4 could never do. Also the ability to export and import the "memory" will be incredibly helpful. Tbh I'm still a bit shocked on how much of a difference prompt…

Hey, thanks, glad you checked that out! I’ve got an update brewing with native ctags support (rather than uploading a binary) that ties better into the symbol tree, and a multi-turn mode for longer edits.

Re: ChatGPT’s system prompts

#319

Post by desolved »

mritchie712 wrote: post #79
simonw wrote: post #6

I 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…

In the training data, it's likely more common to see: "Please [do something]" Then it is to see: "You must [do something]" "Please" makes it clear that what comes next is a command to do something.

I think this is correct. I read somewhere the prompt:

Can you [do something]

Is inferior to:

[do something]

Re: ChatGPT’s system prompts

#320

Post by spdustin »

adeelk93 wrote: post #308
spdustin wrote: post #154

Earlier quoted context omitted.

That was one of several methods, but I (usually) don’t go that deep in comment threads. For Advanced Data Analysis, I had it “use Jupyter to write Python” to transform the content of our conversation, including “messages that appeared before this one” or “after ‘You are ChatGPT’”, into a list of dicts. For both voice and mobile, I opened the same Advanced Data Analysis chat in the iOS client, pointed out that I belie…

I am a bit suspicious because ChatGPT knows which version of Python it has installed, as well as which packages, without actually executing any Python. And yet, this context is not in the shared system prompt for advanced data analysis.

Watch the networking panel in devtools. I don’t think it’s a different tuning, per se, but I do think there’s a middleware, and network traffic suggests that’s what the middleware handles.
Post reply on HN