One does not train or program GPT-3 in a normal way, but one engages in dialogue and writes prompts to teach GPT-3 what one wants.
They demonstrate an ability to handle abstractions, like style parodies, I have not seen in GPT-2 at all. Chatting with GPT-3 feels uncannily like chatting with a human. I was impressed by the results reported in the GPT-3 paper, and after spending a week trying it out, I remain impressed.
I hope you enjoy them even a tenth as much as I enjoyed testing GPT-3 and watching the completions scroll across my screen. Scaling works: quantity is a quality all its own. The scaling of GPT What can we do with GPT-3? Must we content ourselves with mediocre generic poetry, at best, deprived of finetuning directly on chosen poetry corpuses or authors we might like to parody?
How much does GPT-3 improve and what can it do? Turns out: a lot! Below, I walk through first impressions of using GPT-3, and countless samples. You can skip to the appendix for more example like its poems , or browse the random samples.
Seuss poems a particular highlight. Merzmensch Kosmopol enjoyed generating love letters written by a toaster. Tomer Ullman prompted GPT-3 for new philosophy thought experiments. Harley Turan found that, somehow, GPT-3 can associate plausible color hex codes with specific emoji apparently language models can learn color from language, much like blind humans do.
