Update app.py

app.py

@@ -37,7 +37,7 @@ model_name = "quantumaikr/llama-2-70b-fb16-korean"
 #daekeun-ml/Llama-2-ko-DPO-13B
 #daekeun-ml/Llama-2-ko-instruct-13B
 #quantumaikr/llama-2-70b-fb16-korean
-tokenizer = AutoTokenizer.from_pretrained(model_name, use_fast=False)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
 
 model = None
 model = AutoDistributedModelForCausalLM.from_pretrained(model_name)
@@ -263,8 +263,8 @@ def chat(id, npc, text):
 [대화기록]{history[npc][id]}
 """
 
-        #inputs = tokenizer("아", return_tensors="pt")["input_ids"]
-        #outputs = model.generate(inputs, do_sample=True, temperature=0.6, top_p=0.75, max_new_tokens=2)
+        inputs = tokenizer("아", return_tensors="pt")["input_ids"]
+        outputs = model.generate(inputs, do_sample=True, temperature=0.6, top_p=0.75, max_new_tokens=2)
         #output = tokenizer.decode(outputs[0])[len(prom)+3:-1].split("<")[0].split("###")[0].replace(". ", ".\n")
         #output = cleanText(output)
         #print(tokenizer.decode(outputs[0]))