[BUGFIX] KeyError 'layers.14.mlp.gate.g_idx' for Qwen3-MoE with GPTQ on ROCm (#22017)
This commit is contained in:
@ -149,7 +149,7 @@ class Qwen3MoeSparseMoeBlock(nn.Module):
|
||||
self.gate = ReplicatedLinear(config.hidden_size,
|
||||
config.num_experts,
|
||||
bias=False,
|
||||
quant_config=None,
|
||||
quant_config=quant_config,
|
||||
prefix=f"{prefix}.gate")
|
||||
|
||||
def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
|
||||
|
||||
Reference in New Issue
Block a user