News for hackers. Stuff that matters.

A comment on Don Knuth Plays with ChatGPT. Parent comment

Re:Don Knuth Plays with ChatGPT by dchest on Saturday May 20, 2023 @12:28PM (#36013276)
It's not for moral reasons, the reason is simple and practical: ChatGPT is modeling conversations; to get better results, the conversation should look like what it has seen in the training data.
[ Parent | on HN ]
Re:Don Knuth Plays with ChatGPT by coffeebeqn on Saturday May 20, 2023 @12:43PM (#36013414)
The prompt is very important but I don’t think having a polite conversation is usually the best approach. I find giving a briefing with bullet points and ideally an example is much better. There is limited context so you shouldn’t waste it on pretending that you’re talking to a human
[ Parent | on HN ]
Re:Don Knuth Plays with ChatGPT by ianbutler on Saturday May 20, 2023 @12:45PM (#36013443)
This is the right answer, I just casually grepped through a few datasets for instruction tuning I have lying around and please is sprinkled all throughout them.
[ Parent | on HN ]
Re:Don Knuth Plays with ChatGPT by SoftTalker on Saturday May 20, 2023 @03:10PM (#36014685)
Does ChatGPT continually learn from its ongoing conversations? Or is it only trained in advance?
[ Parent | on HN ]