improve score

app.py

@@ -15,9 +15,7 @@ INTRODUCTION_TEXT = """
 - This test is for 15 years old Malaysia student, it is about reading comprehension and general knowledge for malay language.
 2. Tatabahasa, contains 349 questions, https://github.com/mesolitica/malaysian-dataset/tree/master/llm-benchmark/tatabahasabm.tripod.com
 - This test is general test for malay grammar.
-3. Translated IndoNLI to Malay, tested on `test_expert` dataset, https://huggingface.co/datasets/mesolitica/translated-indonli
-- This test is general test to language reasoning.
-4. HumanEval, https://github.com/openai/human-eval
+3. HumanEval, https://github.com/openai/human-eval
 - This test is for programming language understanding.
 """
 
@@ -71,10 +69,13 @@ open_source = [
         'Tatabahasa 3-shots': 27.507163323782237,
     },
     {
-        'model': '[malaysian-llama2-7b-32k-instructions](https://huggingface.co/mesolitica/malaysian-llama2-7b-32k-instructions)',
-        'BM-PT3 0-shot': 35.294117647058826,
-        'BM-PT3 1-shot': 21.153846153846153,
-        'BM-PT3 3-shots': 28.30188679245283,
+        'model': '[malaysian-llama2-7b-32k-instructions](https://huggingface.co/mesolitica/malaysian-llama2-7b-32k-instructions-v2)',
+        'BM-PT3 0-shot': 33.33333333333333,
+        'BM-PT3 1-shot': 37.03703703703704,
+        'BM-PT3 3-shots': 35.18518518518518,
+        'Tatabahasa 0-shot': 54.72779369627507,
+        'Tatabahasa 1-shot': 48.42406876790831,
+        'Tatabahasa 3-shots': 41.833810888252145,
     },
     {
         'model': '[malaysian-llama2-13b-32k](https://huggingface.co/mesolitica/llama-13b-hf-32768-fpf)',
@@ -120,7 +121,7 @@ open_source = [
         'BM-PT3 0-shot': 40.74074074074074,
         'BM-PT3 1-shot': 31.48148148148148,
         'BM-PT3 3-shots': 24.074074074074073,
-        'Tatabahasa 0-shot': 56.446991404011456,
+        'Tatabahasa 0-shot': 57.879656160458445,
         'Tatabahasa 1-shot': 49.28366762177651,
         'Tatabahasa 3-shots': 53.86819484240688
     },