Update app.py

app.py

@@ -88,14 +88,16 @@ def load_model_and_tokenizer(doc_text_count):
     # Load the pre-trained model with quantization
     model = AutoModelForCausalLM.from_pretrained(
         model_name,
-        device_map="auto",  # Automatically allocate model to devices
-        model_max_length=1.3*round(doc_text_count * 0.3 + 1),
-        token=HF_TOKEN
+        trust_remote_code=True,
+        model_max_length=1.3*round(doc_text_count * 0.3 + 1)
+        ,token=HF_TOKEN
     )
 
     # Load the tokenizer
     tokenizer = AutoTokenizer.from_pretrained(
-        model_name
+        model_name,
+        trust_remote_code=True,
+        model_max_length=1.3*round(doc_text_count * 0.3 + 1)
         ,token=HF_TOKEN
     )
     return tokenizer, model