You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on May 29, 2026. It is now read-only.
padding embedding in my mind is usually 1 shared token, but GLIDE use nn.Parameter. My question is what about 128K context-window?Why not share padding token?
padding embedding in my mind is usually 1 shared token, but GLIDE use nn.Parameter. My question is
what about 128K context-window?Why not share padding token?Thanks for your attention
code is at https://github.com/openai/glide-text2im/blob/main/glide_text2im/text2im_model.py#L10