Reduced tockens

app.py

@@ -4,7 +4,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 import re
 
 # Load model and tokenizer
-print("Loading DialoGPT-medium...")
+print("Loading gemma...")
 tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b-it")
 model = AutoModelForCausalLM.from_pretrained("google/gemma-2b-it")
 
@@ -107,7 +107,7 @@ def respond(message, history):
         with torch.no_grad():
             chat_history_ids = model.generate(
                 input_ids,
-                max_length=1250,
+                max_length=250,
                 temperature=0.9,
                 top_p=0.9,
                 repetition_penalty=1.2,