Update app.py

app.py

@@ -74,7 +74,7 @@ def chat_inf(system_prompt,prompt,history,memory,client_choice,seed,temp,tokens,
     print("\n######### HIST "+str(in_len))
     print("\n######### TOKENS "+str(tokens))
     if (in_len+tokens) > 8000:
-        history.append((prompt,"Wait, that's too many tokens, please reduce the Chat Memory value"))
+        history.append((prompt,"Wait, that's too many tokens, please reduce the 'Chat Memory' value, or reduce the 'Max new tokens' value"))
         yield history,memory
         #hist=compress_history(history,client_choice,seed,temp,tokens,top_p,rep_p,chat_mem)
         #yield [(prompt,"History has been compressed, processing request...")]
@@ -145,11 +145,11 @@ with gr.Blocks() as app:
                 with gr.Group():
                     rand = gr.Checkbox(label="Random Seed", value=True)
                     seed=gr.Slider(label="Seed", minimum=1, maximum=1111111111111111,step=1, value=rand_val)
-                    tokens = gr.Slider(label="Max new tokens",value=3840,minimum=0,maximum=8000,step=64,interactive=True, visible=True,info="The maximum number of tokens")
+                    tokens = gr.Slider(label="Max new tokens",value=1600,minimum=0,maximum=8000,step=64,interactive=True, visible=True,info="The maximum number of tokens")
                     temp=gr.Slider(label="Temperature",step=0.01, minimum=0.01, maximum=1.0, value=0.9)
                     top_p=gr.Slider(label="Top-P",step=0.01, minimum=0.01, maximum=1.0, value=0.9)
                     rep_p=gr.Slider(label="Repetition Penalty",step=0.1, minimum=0.1, maximum=2.0, value=1.0)
-                    chat_mem=gr.Number(label="Chat Memory", info="Number of previous chats to retain",value=5)
+                    chat_mem=gr.Number(label="Chat Memory", info="Number of previous chats to retain",value=4)
         with gr.Accordion(label="Screenshot",open=False):
             with gr.Row():
                 with gr.Column(scale=3):