Update app.py

app.py

@@ -14,7 +14,7 @@ def format_prompt(system, message, history):
   return prompt
 
 def generate(
-    prompt, history, system_prompt, temperature=0.8, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
+    prompt, history, system_prompt, temperature=0.8, max_new_tokens=128, top_p=0.95, repetition_penalty=1.0,
 ):
     temperature = float(temperature)
     if temperature < 1e-2:
@@ -50,7 +50,7 @@ additional_inputs=[
     ),
     gr.Slider(
         label="Temperature",
-        value=0.9,
+        value=0.8,
         minimum=0.0,
         maximum=1.0,
         step=0.05,
@@ -59,16 +59,16 @@ additional_inputs=[
     ),
     gr.Slider(
         label="Max new tokens",
-        value=256,
+        value=128,
         minimum=0,
-        maximum=1048,
+        maximum=250,
         step=64,
         interactive=True,
         info="The maximum numbers of new tokens",
     ),
     gr.Slider(
         label="Top-p (nucleus sampling)",
-        value=0.90,
+        value=0.95,
         minimum=0.0,
         maximum=1,
         step=0.05,
@@ -77,7 +77,7 @@ additional_inputs=[
     ),
     gr.Slider(
         label="Repetition penalty",
-        value=1.2,
+        value=1.0,
         minimum=1.0,
         maximum=2.0,
         step=0.05,
@@ -100,5 +100,5 @@ gr.ChatInterface(
     additional_inputs=additional_inputs,
     title="Dolphin 2.8 Gemma 2b",
     examples=examples,
-    concurrency_limit=20,
+    concurrency_limit=5,
 ).launch(show_api=False)