-
-
Notifications
You must be signed in to change notification settings - Fork 4.4k
Open
Labels
Description
Hi,
I am using WSL2+Ubuntu 24, RTX 5090. I am running the given code. but at the before inference cell, I'm getting this error, even though i haven't changed anything.
"name": "ValueError",
"message": "block_mask was created for block_mask.shape=(1, 1, 39, 39) but got q_len=1 and kv_len=39. As the block mask was created for a larger length than you're using it for, you can either 1. create a new block mask with the correct length, or 2. 'adjust' the existing block mask to the correct length by calling block_mask._adjust(q_len, kv_len). This essentially 'crops' the block mask to the upper left corner, which does not work for all mask_mods!",
Reactions are currently unavailable