fix: Initialize ApertusMLP's xielu activation using `torch_dtype` #42864
wasertech
force pushed
from
91f68a81
to
67306014
14 days ago
wasertech
force pushed
from
67306014
to
26667d38
14 days ago
wasertech
force pushed
from
26667d38
to
1cb8b00b
14 days ago
Fix Apertus model crash on float16 hardware
4b41cf89
wasertech
force pushed
from
1cb8b00b
to
4b41cf89
14 days ago
refactor: Move `ACT2CLS` import to top-level in Apertus models.
a660db51
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub