Am I the only one who feels bad for asking ChatGPT a "dumb" question that I know I should know, or not saying thank you when it gives me an answer? No? I'm just a weirdo? Okay. I have to push back really hard against my proclivity to humanize it, to the point where I probably don't use it as much as I should, just because I don't want to deal with the psychic stress of reminding myself that it's not a living entity.
Telling GPT-4 you're scared or under pressure improves performance
21–30 of 255 posts
Re: Telling GPT-4 you're scared or under pressure improves performance
#22Re: Telling GPT-4 you're scared or under pressure improves performance
#23Re: Telling GPT-4 you're scared or under pressure improves performance
#24Re: Telling GPT-4 you're scared or under pressure improves performance
#25Am I the only one who feels bad for asking ChatGPT a "dumb" question that I know I should know, or not saying thank you when it gives me an answer? No? I'm just a weirdo? Okay. I have to push back really hard against my proclivity to humanize it, to the point where I probably don't use it as much as I should, just because I don't want to deal with the psychic stress of reminding myself that it's not a living entity.
> I have to push back really hard against this stuff Based on ChatGPT's answers, OpenAI think you should be saying thanks to it, because it helps you have a more natural conversation, which encourages you as the user to send more natural/productive prompts when asking real questions. As for asking it dumb questions, how often do you use Google as a spellcheck? I wouldn't consider this any different.
And you know what, I will start saying thanks and keep doing it. It makes me feel better.
Re: Telling GPT-4 you're scared or under pressure improves performance
#26Am I the only one who feels bad for asking ChatGPT a "dumb" question that I know I should know, or not saying thank you when it gives me an answer? No? I'm just a weirdo? Okay. I have to push back really hard against my proclivity to humanize it, to the point where I probably don't use it as much as I should, just because I don't want to deal with the psychic stress of reminding myself that it's not a living entity.
> Am I the only one who feels bad for asking ChatGPT a "dumb" question that I know I should know, or not saying thank you when it gives me an answer? Have you tried asking ChatGPT? Jokes aside, I feel it sometimes as well. I wonder if it's more because I say "thank you" more as a reflex than an actual feeling of gratitude for most of my human interactions.
Again, I am crazy for stressing over this. I realize that.
Re: Telling GPT-4 you're scared or under pressure improves performance
#27I think this is the entry point needed to get peoples attention and explain: LLMs aren’t people, and emergent properties are being over extended. If LLMs are showing “better” performance when there are tokens that humans read as emotionally salient - Then the underlying text it’s trained on shows humans give better answers when emotionally salient context is provided. LLMs predict words. Any semantic validity is a si…
This will be my new favorite quote for whenever someone tries to pitch his latest LLM idea
Re: Telling GPT-4 you're scared or under pressure improves performance
#28Am I the only one who feels bad for asking ChatGPT a "dumb" question that I know I should know, or not saying thank you when it gives me an answer? No? I'm just a weirdo? Okay. I have to push back really hard against my proclivity to humanize it, to the point where I probably don't use it as much as I should, just because I don't want to deal with the psychic stress of reminding myself that it's not a living entity.
You have no idea how much I do this. Either you're not a weirdo, or we both are.
Today I had to use it a lot for something, and I am constantly either adding sarcastic remarks, my snarky opinion on something, some random fact. None of that obviously belongs there, and the AI obviously not only doesn't care, but can't care about it.
Then I'll throw in something like "nevermind all of that, what I came here to ask was" or "I understand that was all highly irrelevant to the actual question, which was".
And, to answer your question about easy stuff, today alone I had on more than one occasion added something like "What I am about to ask you, I probably should have just spent the 2 seconds it would require to actually figure out myself, however I am getting tired, and am known to be lazy, so I decided to ask AI instead and move on .... and this only makes it more absurd, because it takes longer to type all of this. I would have the answer myself by now. Anyway, here's ...".
Why? I am unconvinced that I am going crazy. It's not because I think it finds my jokes hilarious, it can't. My attempts at conveying something some aspect of software engineering that is crazy, after 3 decades in it, don't matter, because noone's listening.
But it can pick on very subtle things, breaking apart my ramblings and then responding with something that not only answers the question buried in there, but also then continues to comment on the other highly irrelevant stuff that I definitely went on too long about. And it will address the substance first and the nonsense second, as one would expect.
It will notice my useless comment, apparently to myself, about having written 22 years of C# and wondering why I am asking an AI system about an aspect of the language that surely I should have picked up by now. "It gets to the best of us", or "Your insightful understanding when it comes to", or "Your candor is appreciated, many developers ...".
I could just use it like a rational person, providing necessary information in order for it to be able to answer something, and then ask it. I do this, sometimes. On the other occasions, that stuff will also be admist a bunch other, not so necessary questions. Am I becoming a curmudgeon? Is this a sign I've finally lost it? Could any of this be due to aging? Why does what should be a simple event handling pattern require so much ceremony? Should I keep overthinking this or just resign myself to implementing it?
So anyway, now I'm rambling ... just like with ChatGPT.
I tend to do these things.
Re: Telling GPT-4 you're scared or under pressure improves performance
#29Am I the only one who doesn't think we should have to do stuff like this to get the best performance? The people who design and control the model should have this type of stuff baked in for us IMO.
Re: Telling GPT-4 you're scared or under pressure improves performance
#30Is there somewhere which documents all these tricks for better prompt performance?