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
@@ -1684,8 +1684,6 @@ class DeepseekV2ForCausalLM(
self.set_moe_parameters()
def set_moe_parameters(self):
self.expert_weights = []
self.num_expert_groups = getattr(self.config, "n_group", 1)
self.moe_layers = []