microsoft / unilm

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities
https://aka.ms/GeneralAI
MIT License
19.59k stars 2.5k forks source link

BeiT3 giant model weights release #1031

Open runzeer opened 1 year ago

runzeer commented 1 year ago

Thanks for your code release for the BeiT3. But the model in your paper said the layers for the BeiT is 40. But I saw that the BeiT large is only 24 layers. So any plans for the BeiT3 giant model weights release?

nathanodle commented 1 year ago

@addf400

thesby commented 3 months ago

Could you please release giant model?