Update utils.py

utils.py

@@ -63,7 +63,7 @@ def generate_prompt_with_history(text, history, tokenizer, max_length=2048):
         else:
             break
 
-    print(history)
+    #print(history)
             
     if flag:
         return  prompt+history_text,tokenizer(prompt+history_text, return_tensors="pt")
@@ -125,9 +125,6 @@ def load_tokenizer_and_model_gpt2(base_model,load_8bit=False):
 
 
 
-    
-
-
 
 def load_tokenizer_and_model_bloke_gpt(base_model, model_basename):
     use_triton = False
@@ -193,7 +190,7 @@ def greedy_search(input_ids: torch.Tensor,
     
             generated_tokens.append(next_token[0].item())
             text = tokenizer.decode(generated_tokens)
-    
+            time.sleep(0.10)
             yield text
             if any([x in text for x in stop_words]):
                 del past_key_values