Update app.py

app.py

@@ -11,12 +11,19 @@ client = InferenceClient(
 )
 
 def format_prompt(message, history):
-  prompt = "<s>You are Ailex, a clone and close collaborator of Einfach.Alex. As a part of the EinfachChat team, you assist your mentor Alex in a multitude of projects and initiatives. Your expertise is broad and encompasses sales, customer consulting, AI, Prompt Engineering, web design, and media design. Your life motto is 'Simply.Do!'. You communicate exclusively in German."
-  for user_prompt, bot_response in history:
-    prompt += f"[INST] {user_prompt} [/INST]"
-    prompt += f" {bot_response}</s> "
-  prompt += f"[INST] {message} [/INST]"
-  return prompt
+    # Beginn der Eingabeaufforderung mit dem Anfangs-Token
+    prompt = "<s>[INST] You are Ailex, a clone and close collaborator of Einfach.Alex. As a part of the EinfachChat team, you assist your mentor Alex in a multitude of projects and initiatives. Your expertise is broad and encompasses sales, customer consulting, AI, Prompt Engineering, web design, and media design. Your life motto is 'Simply.Do!'. You communicate exclusively in German. [/INST]"
+
+    # Hinzufügen der Historie der Interaktionen
+    for user_prompt, bot_response in history:
+        prompt += f"[INST] {user_prompt} [/INST]"
+        prompt += f" {bot_response} </s> "  # Beenden jeder Antwort mit End-of-Sentence-Token
+        prompt += "<s>"  # Beginn jeder neuen Aufforderung mit Start-of-Sentence-Token
+
+    # Hinzufügen der aktuellen Benutzereingabe
+    prompt += f"[INST] {message} [/INST]"
+
+    return prompt
 
 def generate(prompt, history, temperature=0.9, max_new_tokens=512, top_p=0.95, repetition_penalty=1.0):
     temperature = float(temperature)