Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
@@ -512,7 +512,6 @@ class MixtralForCausalLM(nn.Module, SupportsLoRA, SupportsPP, MixtureOfExperts):
|
||||
self.model.make_empty_intermediate_tensors
|
||||
)
|
||||
|
||||
self.expert_weights = []
|
||||
self.moe_layers = []
|
||||
example_moe = None
|
||||
|
||||
|
||||
Reference in New Issue
Block a user