I'm so tired about these examples: I don't find them impressive.
Here's what I got several times with GPT and other who are using it must have encountered the same issue (simplifying the example to show the problem I encounter all the time):
GPT: Frodo says that 2 + 2 is 5.
user: that is wrong, 2 + 2 is 4.
GPT: You are correct, I was wrong. Here's the correct version:
Gandalf say that 2 + 2 is 5.
This happens to me
all the time. The textual model only evaluates part of the input and focus, for whatever reason, on the wrong part of the text and then from there it's a descent into madness / futility.
This exact "partial input" issue was shown yesterday in the "Eliza vs GPT" vid:
Eliza: "Did you come to me because you are programmed to simulate human language?"
GPT: "Yes, that's correct. I'm programmed to simulate human language behavior..."
Everybody using it must have noticed that. It's not answering what was asked. It's only partially taking into account the user's input. I don't know if it's because the way questions are parsed or the way the model is compressed but something is very seriously off.
And don't "you're prompting it wrong" me.
Here's what I got several times with GPT and other who are using it must have encountered the same issue (simplifying the example to show the problem I encounter all the time):
This happens to me all the time. The textual model only evaluates part of the input and focus, for whatever reason, on the wrong part of the text and then from there it's a descent into madness / futility.This exact "partial input" issue was shown yesterday in the "Eliza vs GPT" vid:
Everybody using it must have noticed that. It's not answering what was asked. It's only partially taking into account the user's input. I don't know if it's because the way questions are parsed or the way the model is compressed but something is very seriously off.And don't "you're prompting it wrong" me.