fix: disable fast tokenizer to avoid SentencePiece parse error

app.py

@@ -8,7 +8,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
 
 # ✅ Public, compatible model
 model_id = "NousResearch/Nous-Hermes-2-Mistral-7B-DPO"
-tokenizer = AutoTokenizer.from_pretrained(model_id)
+tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=False)
 model = AutoModelForCausalLM.from_pretrained(model_id)
 
 llm_pipeline = pipeline("text-generation", model=model, tokenizer=tokenizer, max_new_tokens=512)