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:
Harry Mellor
2026-06-27 00:16:05 +01:00
committed by GitHub
parent 2ff76a5e85
commit d8eb734d94
29 changed files with 80 additions and 156 deletions
-2
View File
@@ -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