Add GPTQ support (#916)

vllm/model_executor/models/chatglm.py

@@ -377,6 +377,9 @@ class ChatGLMForCausalLM(nn.Module):
                 continue
             if "word_embeddings" in name:
                 name = name.replace(".word_embeddings", "")
+            # Skip loading extra bias for GPTQ models.
+            if name.endswith(".bias") and name not in params_dict:
+                continue
             param = params_dict[name]
             weight_loader = getattr(param, "weight_loader",
                                     default_weight_loader)