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
ValueError: Incompatible shapes for attention inputs:
query.shape: torch.Size([64, 4096, 40])
key.shape : torch.Size([32, 77, 40])
value.shape: torch.Size([32, 77, 40])
HINT: We don't support broadcasting, please use `expand` yourself before calling `memory_efficient_attention` if you need to
tried without xFormers and got:
RuntimeError: The size of tensor a (8192) must match the size of tensor b (4096) at non-singleton dimension 1
How can I prevent this ?
The text was updated successfully, but these errors were encountered:
When I use
stream.update_prompt(prompt)
I get :
tried without xFormers and got:
RuntimeError: The size of tensor a (8192) must match the size of tensor b (4096) at non-singleton dimension 1
How can I prevent this ?
The text was updated successfully, but these errors were encountered: