The projection follows rapid capability gains seen from GPT-3 to Fable.
The projection follows rapid capability gains seen from GPT-3 to Fable.
@paulg Wow, this is absolutely amazing! It feels like something straight out of a sci-fi movie! 🚀✨
@paulg Fable is retarded.
@paulg So not much?
Where much of the field saw flaws - Gwern understood that the fact it worked so well despite being so flawed meant unbelievable headroom. Models in 5 years will be as Fable to GPT3 - every part of what we do today could be radically better! https://x.com/_sholtodouglas/status/2074234813146697769/photo/1
A good time to reflect on https://gwern.net/scaling-hypothesis. It took two more years for people to begin waking up. https://x.com/_sholtodouglas/status/2074234810755895578/photo/1 https://twitter.com/paulg/status/2074214198712680947
Imagine what it will be like if 5 years from now models have improved on Fable as much as Fable has improved on GPT3. https://twitter.com/snowmaker/status/2073948669536260163
And are run locally on your phone. https://twitter.com/paulg/status/2074214198712680947
@alexolegimas My phone will tell me off regularly in 5 years?
@_sholtodouglas @inductionheads the race is on mr sholto.
Ask a question below.
Published answers will appear here.
