Update README.md

README.md

@@ -44,7 +44,7 @@ tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm3-6b-base", trust_remote
 model = AutoModel.from_pretrained("THUDM/chatglm3-6b-base", trust_remote_code=True).half().cuda()
 
 inputs = tokenizer(["今天天气真不错"], return_tensors="pt").to('cuda')
-outputs = tokenizer.generate(**inputs)
+outputs = model.generate(**inputs)
 print(tokenizer.decode(outputs[0].tolist()))
 ```
 
@@ -77,4 +77,4 @@ For more instructions, including how to run CLI and web demos, and model quantiz
   pages={320--335},
   year={2022}
 }
-```
+```