News for hackers. Stuff that matters.

A comment on Anthropic announces Claude 2.1 – 200k context, less refusals. Parent comment

Re:Anthropic announces Claude 2.1 – 200k context, less refusals by diplodinkus on Tuesday November 21, 2023 @11:57AM (#38366325)
Agreed, but I do find gpt4 has been increasing the amount of pseudo code recently. I think they are a/b testing me. I find myself asking if how much energy it wasted giving me replies that I then have to tell it to fix.. Which is of course a silly thing to do, but maybe someone at oAI is listening?
[ Parent | on HN ]
Re:Anthropic announces Claude 2.1 – 200k context, less refusals by FrenchDevRemote on Tuesday November 21, 2023 @11:59AM (#38366350)
If you mean through the user friendly chat GPT website, they're probably making it output as few tokens as possible to cut costs
[ Parent | on HN ]
Re:Anthropic announces Claude 2.1 – 200k context, less refusals by FrustratedMonky on Tuesday November 21, 2023 @12:07PM (#38366486)
That can't be, because I can ask it a simple question that an answer is maybe 1 sentence, and it repeats the question then provides a whole novel. So ton of tokens.
[ Parent | on HN ]
Re:Anthropic announces Claude 2.1 – 200k context, less refusals by madeofpalk on Tuesday November 21, 2023 @12:12PM (#38366565)
GPT still writes like a highschooler trying to hit a high word count :(
[ Parent | on HN ]
Re:Anthropic announces Claude 2.1 – 200k context, less refusals by droopyEyelids on Tuesday November 21, 2023 @12:52PM (#38367191)
Like a content mill trying to keep you on the page for as long as possible! Which it was trained on.
[ Parent | on HN ]
Re:Anthropic announces Claude 2.1 – 200k context, less refusals by gtirloni on Tuesday November 21, 2023 @01:08PM (#38367483)
You can ask it to be very concise.

I added it to my custom instructions and it has helped a lot.

[ Parent | on HN ]
Re:Anthropic announces Claude 2.1 – 200k context, less refusals by gumballindie on Tuesday November 21, 2023 @12:08PM (#38366504)
Wow, imagine paying so they can experiment on you and limit what you get. I so wish i found such … useful clients for my own projects.
[ Parent | on HN ]
Re:Anthropic announces Claude 2.1 – 200k context, less refusals by FrenchDevRemote on Wednesday November 22, 2023 @02:46AM (#38376206)
It's not experimentation, it's probably one of the only things that allowed them to make gpt 3.5 turbo 10 TIMES cheaper than the previous model.
[ Parent | on HN ]