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:
@@ -655,8 +655,6 @@ class Glm4MoeForCausalLM(nn.Module, SupportsPP, SupportsLoRA, Glm4MixtureOfExper
|
||||
self.make_empty_intermediate_tensors = (
|
||||
self.model.make_empty_intermediate_tensors
|
||||
)
|
||||
self.expert_weights = []
|
||||
|
||||
# Set MoE hyperparameters
|
||||
self.num_moe_layers = config.num_hidden_layers - config.first_k_dense_replace
|
||||
self.num_expert_groups = config.n_group
|
||||
|
||||
Reference in New Issue
Block a user