diff --git a/chinese-wiki.py b/chinese-wiki.py
index e0afb99..dc979a7 100644
--- a/chinese-wiki.py
+++ b/chinese-wiki.py
@@ -2,6 +2,7 @@ from utils import process_dataset
 
 if __name__ == "__main__":
     process_dataset(
-        dataset_name="Blaze7451/Wiki-zh-20250601",
-        output_subdir="chinese-wiki"
+        dataset_name="wikimedia/wikipedia",
+        output_subdir="chinese-wiki",
+        dataset_config="zh"
     )
\ No newline at end of file
diff --git a/utils.py b/utils.py
index 1e15b6a..610ab00 100644
--- a/utils.py
+++ b/utils.py
@@ -24,7 +24,7 @@ def process_dataset(
     normalization_func=comprehensive_normalization
 ):
 
-    dataset_dict = load_dataset(dataset_name, dataset_config) if dataset_config else load_dataset(dataset_name)
+    dataset_dict = load_dataset(dataset_name, dataset_config)
     train_dataset = dataset_dict[split_name]
 
     total_samples = len(train_dataset)