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

New GPT-3 model: text-DaVinci-003

beta.openai.com

Post reply on HN
Print view View original

11–20 of 216 posts

Re: New GPT-3 model: text-DaVinci-003

#11

Post by zoba »

Oh wow! It can rhyme!

  Write a rhyming poem comparing love to a beautiful flower:

  A flower so beautiful and fair

  Love can be found everywhere

  Its petals so vibrant and bright

  A love that will last day and night

  The beauty of love, like a flower

  It's a beauty that will never cower

  A feeling of joy that will last

  Like a flower, love stands fast

  The beauty of love is so true

  Like a blooming flower for me and for you

  No matter the season, it will stay

  Love is like a beautiful flower, in every way.

Re: New GPT-3 model: text-DaVinci-003

#12

Post by faizshah »

ctoth wrote: post #3

This one seems weirdly more verbose, in my brief testing. Definitely uses more tokens on average than 002 when answering. The writing style seems oddly-pretentious and yet somehow uninformed, perhaps there were too many copies of HN in the dataset? I kid, I kid.

It can't have been trained on HN, I tried generating my comment for this thread and it would only output positive marketing speak instead of "This name seems to collide with the artist Da Vinci."

Re: New GPT-3 model: text-DaVinci-003

#13

Post by AnimalMuppet »

jerpint wrote: post #8

Since it’s trained on more recent scrapes, does this mean that this new version will be self-aware? So if you ask it about “gpt-3” it’ll know what you’re talking about?

That's not what most of us consider "self-aware". That's just knowing your name.

To me, "self-aware" is essentially being able to watch yourself think. I'm pretty sure GPT-3 can't do that, in any model.

Re: New GPT-3 model: text-DaVinci-003

#14

Post by thinkmcfly »

zoba wrote: post #11

Oh wow! It can rhyme! Write a rhyming poem comparing love to a beautiful flower: A flower so beautiful and fair Love can be found everywhere Its petals so vibrant and bright A love that will last day and night The beauty of love, like a flower It's a beauty that will never cower A feeling of joy that will last Like a flower, love stands fast The beauty of love is so true Like a blooming flower for me and for you No m…

002 could rhyme alright with the right prompt

Re: New GPT-3 model: text-DaVinci-003

#16

Post by zoba »

thinkmcfly wrote: post #14
zoba wrote: post #11

Oh wow! It can rhyme! Write a rhyming poem comparing love to a beautiful flower: A flower so beautiful and fair Love can be found everywhere Its petals so vibrant and bright A love that will last day and night The beauty of love, like a flower It's a beauty that will never cower A feeling of joy that will last Like a flower, love stands fast The beauty of love is so true Like a blooming flower for me and for you No m…

002 could rhyme alright with the right prompt

I could get 002 to rhyme individual words, but not in a poetry format. Not claiming to be the world's best prompt engineer though.

Re: New GPT-3 model: text-DaVinci-003

#17

Post by oldgradstudent »

zoba wrote: post #11

Oh wow! It can rhyme! Write a rhyming poem comparing love to a beautiful flower: A flower so beautiful and fair Love can be found everywhere Its petals so vibrant and bright A love that will last day and night The beauty of love, like a flower It's a beauty that will never cower A feeling of joy that will last Like a flower, love stands fast The beauty of love is so true Like a blooming flower for me and for you No m…

Have it compose a poem — a poem about a haircut! But lofty, noble, tragic, timeless, full of love, treachery, retribution, quiet heroism in the face of certain doom! Six lines, cleverly rhymed, and every word beginning with the letter S!!

Re: New GPT-3 model: text-DaVinci-003

#18

Post by _jal »

faizshah wrote: post #12
ctoth wrote: post #3

This one seems weirdly more verbose, in my brief testing. Definitely uses more tokens on average than 002 when answering. The writing style seems oddly-pretentious and yet somehow uninformed, perhaps there were too many copies of HN in the dataset? I kid, I kid.

It can't have been trained on HN, I tried generating my comment for this thread and it would only output positive marketing speak instead of "This name seems to collide with the artist Da Vinci."

I knew it wasn't trained on HN when it didn't immediately try to correct my word choice.

Re: New GPT-3 model: text-DaVinci-003

#20

Post by johnfn »

I've been using GPT-3 recently to give me ideas for my nanowrimo novel. I'll just paste in the last couple of paragraphs that I wrote and see what it thinks I should write next. Most of the time, the responses are pretty silly (it gets into loops a lot or doesn't know where to take the story) but a couple of times I've gotten some really good lines out of it.

It kinda feels like copilot, but for writing!

Post reply on HN