Update README.md

README.md

@@ -13,22 +13,6 @@ This model is an int4 model with group_size 128 of [THUDM/chatglm2-6b](https://h
 
 
 
-### INT4 Inference with AutoGPTQ's Kernel
-
-```python
-##pip install auto-gptq[triton] 
-##pip install triton==2.2.0
-from transformers import AutoModel, AutoTokenizer
-quantized_model_dir = "Intel/chatglm2-6b-int4-inc"
-model = AutoModel.from_pretrained(quantized_model_dir,
-                                  device_map="auto",
-                                  trust_remote_code=False,
-                                )
-
-tokenizer = AutoTokenizer.from_pretrained(quantized_model_dir, use_fast=True)
-print(tokenizer.decode(model.generate(**tokenizer("There is a girl who likes adventure,", return_tensors="pt").to(model.device),max_new_tokens=50)[0]))
-```
-
 
 
 ### Evaluate the model