Update app.py

app.py

@@ -19,25 +19,17 @@ def load_model_org():
     return model, tokenizer
 
 def load_model():
-    print("Loading tokenizer...")
-    tokenizer = AutoTokenizer.from_pretrained(
-        MODEL_NAME, 
-        trust_remote_code=True
-    )
-
-    print("Loading configuration...")
-    config = AutoConfig.from_pretrained(MODEL_NAME)
-    # Set the correct vocab size to match your fine-tuned model
-    config.vocab_size = 128256  
-
-    print("Loading model...")
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, trust_remote_code=True)
+    
     model = AutoModelForCausalLM.from_pretrained(
         MODEL_NAME,
-        config=config,  # Use the modified config
         torch_dtype=torch.float16,
         device_map="auto",
         trust_remote_code=True,
+        load_in_8bit=False
     )
+    
+    model.resize_token_embeddings(len(tokenizer))
     return model, tokenizer
 
 # Initialize model and tokenizer