Update app.py

app.py

@@ -12,11 +12,11 @@ description = "Interestingly there simply wasnt a public demo for Tulu, So I mad
 
 os.environ['PYTORCH_CUDA_ALLOC_CONF'] = 'max_split_size_mb:50'
 device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
-model_name = "allenai/tulu-2-dpo-70b"
-tokenizer = AutoTokenizer.from_pretrained("allenai/tulu-2-dpo-70b")
+model_name = "allenai/tulu-2-dpo-13b"
+tokenizer = AutoTokenizer.from_pretrained("allenai/tulu-2-dpo-13b")
 model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, use_flash_attention_2=True, device_map="auto")
 
-class XgenChatBot:
+class TuluChatBot:
     def __init__(self, model, tokenizer, system_message="You are 🌷Tulu, an AI language model created by Tonic-AI. You are a cautious assistant. You carefully follow instructions. You are helpful and harmless and you follow ethical guidelines and promote positive behavior."):
         self.model = model
         self.tokenizer = tokenizer
@@ -47,11 +47,11 @@ class XgenChatBot:
         return response
 
 def gradio_predict(user_message, system_message, max_new_tokens, temperature, top_p, repetition_penalty):
-    Orca_bot.set_system_message(system_message)
-    response = Orca_bot.predict(user_message, temperature, max_new_tokens, top_p, repetition_penalty)
+    Tulu_bot.set_system_message(system_message)
+    response = Tulu_bot.predict(user_message, temperature, max_new_tokens, top_p, repetition_penalty)
     return response
     
-Orca_bot = OrcaChatBot(model, tokenizer)
+Tulu_bot = TuluChatBot(model, tokenizer)
 
 iface = gr.Interface(
     fn=gradio_predict,