Update app.py

app.py

@@ -58,7 +58,8 @@ generation_config = AutoConfig.from_pretrained(
   #  do_sample=True,
   #  stream = streamer,
     top_p=0.95,
-    temperature=0.4
+    temperature=0.4,
+    stream = True
   #  eos_token_id=terminators
 )
 # send additional parameters to model for generation
@@ -155,6 +156,7 @@ def talk(prompt, history):
     print(*messages)
  #   input_ids = tokenizer(*messages)
 
+    print(model.generate(tensor([[    1,   529, 29989,  5205, 29989]])))
     start = time.time()
     NUM_TOKENS=0
     print('-'*4+'Start Generation'+'-'*4)