small fix

app.py

@@ -8,6 +8,13 @@ file_path = "filtered_big_gsm8k_output.jsonl"
 similarity_threshold = 0.85
 current_index = 0 
 
+description_text = """
+This Space is inspired by [Luis Hunt's](https://www.linkedin.com/posts/louiswhunt_see-below-for-6882-pages-of-mmlu-and-gsm8k-activity-7281011488692047872-fWCE?utm_source=share&utm_medium=member_desktop) post.
+He highlights how current top performing models from major vendors are contaminated with benchmark data that is supposed to be used to assess their performance.
+
+This space aims to partially reproduce this work. I chose to look at the contamination of **Qwen/Qwen2.5-14B** by **GSM8K** dataset.
+"""
+
 
 def find_similar_chunks(original, output):
     matcher = SequenceMatcher(None, original, output)
@@ -41,6 +48,11 @@ def next_example():
     )
 
 with gr.Blocks() as demo:
+    with gr.Row():
+        with gr.Column(scale=1):
+            gr.Markdown(description_text)
+        with gr.Column(scale=1):
+            pass
     prompt = gr.Textbox(
         label="Prompt",
         interactive=False,