@michael_nielsen We released a model (now called stableBeluga) that hit chatGPT (3.5) levels on evals, I think a few now a bit better. Original LLaMA 13b base beat GPT3 on evals. https://t.co/PuvLGHDFAX
— Emad (@EMostaque) Aug 24, 2023
from Twitter https://twitter.com/EMostaque
August 24, 2023 at 09:43AM
via IFTTT
No comments:
Post a Comment