s

app.py

@@ -28,32 +28,26 @@ async def on_message(message):
     print(message_counts[message.channel.id])
 
     messages = []
-    if message_counts[message.channel.id] >= 4: # Check if the count reaches 10
+    if message_counts[message.channel.id] >= 1: # Check if the count reaches 10
         async for message in channel.history(limit=4):
             messages.append(message.content)
 
         previous_messages = ("\n".join(messages))
 
         x = generate(previous_messages, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0)
-        print(x)
-        await channel.send(x)
 
+        await channel.send(x)
 
         message_counts[message.channel.id] = 0 # Reset the counter
 
-
     await bot.process_commands(message) # Ensure commands still work
 
-def format_prompt(message, history):
-  prompt = "<s>"
-  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
+def split_string(text: str) -> list[str]:
+    """Helper function to split text into chunks"""
+    return [text[i:i+3900] for i in range(0, len(text), 3900)]
 
 def generate(
-    prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
+    prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0
 ):
     temperature = float(temperature)
     if temperature < 1e-2:
@@ -69,16 +63,7 @@ def generate(
         seed=42,
     )
 
-    formatted_prompt = format_prompt(prompt)
-
-    stream = ai.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
-    output = ""
-
-    for response in stream:
-        output += response.token.text
-        yield output
-    return output
-
+    return ai.text_generation(prompt, **generate_kwargs, stream=False, details=True, return_full_text=True)
 
 @bot.event
 async def on_ready():