fix incorrect output of Qwen2-7B-Instruct-GPTQ-Int4 and Qwen2-7B-Inst… (

#2717)

fix incorrect output of Qwen2-7B-Instruct-GPTQ-Int4 and Qwen2-7B-Instruct-AWQ
ipex kernel provide func like add_bias, so no need add it outside

Signed-off-by: Wang, Yi A <yi.a.wang@intel.com>

server/text_generation_server/layers/awq/quantize/ipex.py

@@ -44,5 +44,4 @@ def __init__(
     def forward(self, x):
         out_shape = x.shape[:-1] + (self.out_features,)
         out = self.woq_linear(x.reshape(-1, x.shape[-1]))
-        out = out + self.bias if self.bias is not None else out
         return out.reshape(out_shape)

server/text_generation_server/layers/gptq/ipex.py

@@ -122,5 +122,4 @@ def pack(self, linear, scales, zeros, g_idx=None):
     def forward(self, x):
         out_shape = x.shape[:-1] + (self.outfeatures,)
         out = self.woq_linear(x.reshape(-1, x.shape[-1]))
-        out = out + self.bias if self.bias is not None else out
         return out.reshape(out_shape)