search

openai / gpt-2

Code for the paper "Language Models are Unsupervised Multitask Learners"
https://openai.com/blog/better-language-models/
Other
22.46k stars 5.51k forks source link

training from scratch #201

Open leejason opened 4 years ago

leejason commented 4 years ago

Will the code for training from scratch be released after the 1.5B model?

senorblasto commented 4 years ago

Ask the 1.5B model, but I heard it knows for sure.

zjm008 commented 4 years ago

Will the code for training from scratch be released after the 1.5B model?

Here's a way to train or finetune. https://github.com/nshepperd/gpt-2

mikolasan commented 4 years ago

I found more info in an old and already closed issue - https://github.com/openai/gpt-2/issues/19. I believe this link here will be helpful for newcomers (like myself). Ideally this repository needs a wiki.