GENERATIVE AI - AI Dynamics

Meeting in San Francisco to discuss generative AI and ChatGPT

By

–

12 February 2023 20h02

I’m in San Francisco for a few days. Hit me up if you’d like to meet and talk about generative AI, cerebral valley, and the profound brilliance & utter stupidity of chatgpt.

→ View original post on X — @willknight

12 February 2023

Live Testing New Bing Chat Performance with Community

By

@dotcsv

–

12 February 2023 19h22

¡ATENCIÓN DIRECTO! ¿Queréis que probemos el nuevo Bing Chat todos juntos, y comprobemos su rendimiento? Veníos YA para el canal de Youtube que empezamos en 10 minutos. ¡No os lo perdáis! Y corred la voz

→ View original post on X — @dotcsv

12 February 2023

Creative AI Art Brilliance Showcased by Tony

By

@pascal_bornet

–

12 February 2023 18h53

Brilliant! 💡🔊

Credit: tonyannnn#creativity #positive #art pic.twitter.com/E37BNfb47r
— Pascal Bornet (@pascal_bornet) 12 février 2023

Brilliant! Credit: tonyannnn
#creativity #positive #art

→ View original post on X — @pascal_bornet

12 February 2023

LLM Collaboration and Chain Rules Innovation

By

@oriolvinyalsml

–

12 February 2023 18h17

The thread above was, of course, written in collaboration with one of our powerful Large Language Models. Long live Chain Rules!

→ View original post on X — @oriolvinyalsml

12 February 2023

Token Probability in Language Models Explained

By

@oriolvinyalsml

–

12 February 2023 18h17

By smaller events here, we refer to the probability of a token, given past tokens, p(c|ab). In probabilistic language modeling, a “token” is a single unit of text, like a word or part of a word. Modern language models consider a vocabulary size of ~100K tokens.

→ View original post on X — @oriolvinyalsml

12 February 2023

Chain Rule Reduces Token Generation Complexity Exponentially

By

@oriolvinyalsml

–

12 February 2023 18h17

To generate a sequence of 1000 tokens requires an insane 100K^1000 = 10^5000 choices. That’s a lot more than the estimated number of atoms in the universe, 10^82! With the chain rule the number of possible choices is "only" 100K * 1000 = 100M, a much more manageable number.

→ View original post on X — @oriolvinyalsml

12 February 2023

DirectoryLoader LangChain loads multiple documents with sources

By

@langchain

–

12 February 2023 17h43

Loading multiple documents and associating the sources is a great idea! And just happens there’s a “DirectoryLoader” which does exactly that https://
langchain.readthedocs.io/en/latest/modu
les/document_loaders/examples/directory_loader.html
…

→ View original post on X — @langchain

12 February 2023

Ghost Cites and ChatGPT: Academic Citation Crisis Worsens

By

@katecrawford

–

12 February 2023 17h33

This is a classic story about an article that was never written yet became widely cited. "Ghost cites" were always problem in academia. But it's about to get much, much worse. This is how Chat GPT summarizes this famously non-existent article:

→ View original post on X — @katecrawford

12 February 2023