Closed marcoaleixo closed 4 years ago
Hey there, thanks for the issue. I'll look into bringing the 774M model onboard. Will keep you updated.
Cheers
Hello there! gpt2-client
is now compatible with the 774M model! Thank you so much for waiting :)
They released a new model - https://openai.com/blog/gpt-2-6-month-follow-up/ And I think we will need TPU to use that model. I'm trying to use the 345M with TPU using Colab, but when I change to TPU the code uses CPU.