Update app.py

app.py

@@ -14,7 +14,13 @@ def format_prompt(message, history):
     prompt += f"[INST] {message} [/INST]"
     return prompt
 
-def generate(prompt, history, temperature=0.9, max_new_tokens=16000, top_p=0.95, repetition_penalty=1.0,):
+def generate(prompt, history, system_prompt, temperature=0.9, max_new_tokens=16000, top_p=0.95, repetition_penalty=1.0,):
+
+        temperature = float(temperature)
+        if temperature < 1e-2:
+            temperature = 1e-2
+        top_p = float(top_p)
+
         generate_kwargs = dict(
         temperature=temperature,
         max_new_tokens=max_new_tokens,
@@ -22,7 +28,8 @@ def generate(prompt, history, temperature=0.9, max_new_tokens=16000, top_p=0.95,
         repetition_penalty=repetition_penalty,
         do_sample=True,
         seed=42,)
-        formatted_prompt = format_prompt(prompt, history)
+    
+        formatted_prompt = format_prompt(f"{system_prompt}, {prompt}", history)
         stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
         output = ""