Unverified Commit 06d0a3d9 authored by Xuchun Shang's avatar Xuchun Shang Committed by GitHub
Browse files

[Bug fix] use correct func path in deepseek (#5496)


Signed-off-by: default avatarXuchun Shang <xuchun.shang@linux.alibaba.com>
parent 22c2a79d
...@@ -170,7 +170,7 @@ class DeepseekMoE(nn.Module): ...@@ -170,7 +170,7 @@ class DeepseekMoE(nn.Module):
shared_output = self.shared_experts(hidden_states) shared_output = self.shared_experts(hidden_states)
# router_logits: (num_tokens, n_experts) # router_logits: (num_tokens, n_experts)
router_logits, _ = self.gate(hidden_states) router_logits, _ = self.gate(hidden_states)
final_hidden_states = fused_moe( final_hidden_states = fused_moe.fused_moe(
hidden_states, hidden_states,
self.w1, self.w1,
self.w2, self.w2,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment