Remove ScaledActivation for AWQ (#10057)

Signed-off-by: mgoin <michael@neuralmagic.com>
2024-11-06 09:27:06 -05:00
parent 406d4cc480
commit 399c798608
34 changed files with 19 additions and 124 deletions
--- a/vllm/model_executor/models/mpt.py
+++ b/vllm/model_executor/models/mpt.py
@@ -153,7 +153,7 @@ class MPTMLP(nn.Module):
            bias=not config.no_bias,
            quant_config=quant_config,
        )
-        self.act = get_act_fn("gelu", quant_config, intermediate_size)
+        self.act = get_act_fn("gelu")
        self.down_proj = RowParallelLinear(
            intermediate_size,
            hidden_size,