Update app.py

app.py

@@ -1,5 +1,5 @@
 import streamlit as st
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from transformers import AutoModel, AutoTokenizer
 from huggingface_hub import login
 import PyPDF2
 import pandas as pd
@@ -80,17 +80,22 @@ def load_model(hf_token):
         
         login(token=hf_token)
         
-        # Load tokenizer
-        tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, token=hf_token)
+        # Load tokenizer with trust_remote_code=True
+        tokenizer = AutoTokenizer.from_pretrained(
+            MODEL_NAME,
+            token=hf_token,
+            trust_remote_code=True
+        )
         
         # Load the model with appropriate dtype for CPU/GPU compatibility
         dtype = torch.float16 if DEVICE == "cuda" else torch.float32
-        model = AutoModelForCausalLM.from_pretrained(
+        model = AutoModel.from_pretrained(
             MODEL_NAME,
             token=hf_token,
             torch_dtype=dtype,
             device_map="auto",  # Automatically maps to CPU or GPU
-            quantization_config=None  # Disable FP8 quantization
+            quantization_config=None,  # Disable FP8 quantization
+            trust_remote_code=True  # Allow custom code execution
         )
         
         return model, tokenizer