added about section for datasets

app.py

@@ -357,6 +357,20 @@ with demo:
                 
         with gr.TabItem("📝 Evaluation Datasets", elem_id="llm-benchmark-tab-table", id=4):
             gr.Markdown(LLM_DATASET_TEXT, elem_classes="markdown-text")
+    with gr.Row():
+        with gr.Column(scale=1):
+            gr.Image('src/display/kapital.jpg', scale=1, 
+                    show_label=False,
+                    interactive=False,
+                    show_share_button=False,
+                    show_download_button=False)
+
+        with gr.Column(scale=1, min_width=1):
+            gr.Image('src/display/kapital.jpg', scale=1, 
+                    show_label=False,
+                    interactive=False,
+                    show_share_button=False,
+                    show_download_button=False)
 
 scheduler = BackgroundScheduler()
 scheduler.add_job(restart_space, "interval", seconds=300)

src/display/about.py

@@ -48,15 +48,32 @@ If you have a fine-tuned Azerbaijani LLM, submit it for evaluation!
 """
 
 LLM_BENCHMARKS_TEXT = f"""
-## How it works
-## Reproducibility
-HERE we have about part
+## Azerbaijani Open LLM sponsored by Kapital Bank
+
+This Open leaderboard 
 """
 
 LLM_DATASET_TEXT = f"""
-## How it works
-## Reproducibility
-HERE we have about part
+##Banking_Call_Classification_MC
+This dataset consists of 192 rows and 4 columns. It is a multiple-choice dataset used to determine which of the presented categories the subject of a request sent to the bank by a client belongs to.
+
+##Banking_Exam_MCQ
+A benchmark dataset of 200-300 multiple-choice questions sourced from universityexam materials across multiple departments,focused specifically on the banking sector in Azerbaijan.
+
+##Banking_QA
+This dataset consists of 97 raws and is a question-answer dataset in the Azerbaijani language about banking.
+
+##Wiki_CQA
+This database consists of 97 rows in Azerbaijani language. It consists of a test consisting of a context from Wikipedia, questions related to that context, and a created answer.
+
+##GSM8K
+A benchmark dataset contain 44 rows,diverse grade school math word problems to measure a model’s abilitiy to solve milti-step mathematical reasoning problems.
+
+##ARC
+This  Benchmark dataset consists of multiple-choice science questions aimed at testing a model's ability to understand and apply elementary scientific knowledge, similar to questions that might appear in standard science exams for students. This version of the dataset is in Azerbaijani, providing an opportunity for models to engage in reasoning and inference in the Azerbaijani language. The dataset is divided into an easy set and a challenge set, with questions requiring reasoning beyond simple fact recall
+
+##Informatics_MC, Azerbaijani_Lang_MC, History_MC, Physics_MC, Geography_MC, LLM-Literature_MC, Logic_MC, Azerbaijani_Hist_MC, Chemistry_MC, Biology_MC
+A comprehensive collection of educational datasets in the Azerbaijani language, covering ten distinct academic disciplines: informatics, Azerbaijani language, world history, physics, geography, literature, logic, Azerbaijani history, chemistry, and biology. Each dataset contains 100 carefully curated multiple-choice questions, designed to assess knowledge and understanding in their respective fields. 
 """