Hao Xu
commited on
Commit
·
e55b0a5
1
Parent(s):
d56bc01
add cc-2025-08 result
Browse files
app.py
CHANGED
|
@@ -33,9 +33,9 @@ def build_table(source, refresh=False):
|
|
| 33 |
data = load_data(source, refresh)
|
| 34 |
|
| 35 |
if source == "core":
|
| 36 |
-
headers = ["Benchmark", "Category", "Pile-train Dirty (%)", "DCLM-baseline Dirty (%)", "CC-2025-05 Dirty (%)"]
|
| 37 |
else:
|
| 38 |
-
headers = ["Benchmark", "Contributor", "Pile-train Dirty (%)", "DCLM-baseline Dirty (%)", "CC-2025-05 Dirty (%)"]
|
| 39 |
|
| 40 |
html = """
|
| 41 |
<table id="benchmarkTable" style="border-collapse: collapse; width: 100%;">
|
|
|
|
| 33 |
data = load_data(source, refresh)
|
| 34 |
|
| 35 |
if source == "core":
|
| 36 |
+
headers = ["Benchmark", "Category", "Pile-train Dirty (%)", "DCLM-baseline Dirty (%)", "CC-2025-05 Dirty (%)", "CC-2025-08 Dirty (%)"]
|
| 37 |
else:
|
| 38 |
+
headers = ["Benchmark", "Contributor", "Pile-train Dirty (%)", "DCLM-baseline Dirty (%)", "CC-2025-05 Dirty (%)", "CC-2025-08 Dirty (%)"]
|
| 39 |
|
| 40 |
html = """
|
| 41 |
<table id="benchmarkTable" style="border-collapse: collapse; width: 100%;">
|
data.json
CHANGED
|
@@ -1,30 +1,30 @@
|
|
| 1 |
[
|
| 2 |
-
{"Benchmark": "MMLU", "Category": "Knowledge and Reasoning", "Pile Dirty": 13.2, "DCLM Dirty": 28.4, "CC202505 Dirty": 13.5, "URL": "https://huggingface.co/datasets/cais/mmlu"},
|
| 3 |
-
{"Benchmark": "MMLU-Pro", "Category": "Knowledge and Reasoning", "Pile Dirty": 5.5, "DCLM Dirty": 16.2, "CC202505 Dirty": 7.1, "URL": "https://huggingface.co/datasets/TIGER-Lab/MMLU-Pro"},
|
| 4 |
-
{"Benchmark": "BBH", "Category": "Knowledge and Reasoning", "Pile Dirty": 0.0, "DCLM Dirty": 0.1, "CC202505 Dirty": 1.4, "URL": "https://github.com/suzgunmirac/BIG-Bench-Hard/tree/main/bbh"},
|
| 5 |
-
{"Benchmark": "AGIEval", "Category": "Knowledge and Reasoning", "Pile Dirty": 0.8, "DCLM Dirty": 3.1, "CC202505 Dirty": 2.7, "URL": "https://github.com/ruixiangcui/AGIEval/tree/main/data/v1_1"},
|
| 6 |
-
{"Benchmark": "GPQA", "Category": "Knowledge and Reasoning", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.89, "URL": "https://huggingface.co/datasets/Idavidrein/gpqa"},
|
| 7 |
-
{"Benchmark": "HLE", "Category": "Knowledge and Reasoning", "Pile Dirty": 0.0, "DCLM Dirty": 0.3, "CC202505 Dirty": 0.1, "URL": "https://huggingface.co/datasets/cais/hle"},
|
| 8 |
|
| 9 |
-
{"Benchmark": "AIME_2024", "Category": "Math", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 10.0, "URL": "https://huggingface.co/datasets/Maxwell-Jia/AIME_2024"},
|
| 10 |
-
{"Benchmark": "GSM8K", "Category": "Math", "Pile Dirty": 0.0, "DCLM Dirty": 0.4, "CC202505 Dirty": 5.0, "URL": "https://huggingface.co/datasets/openai/gsm8k"},
|
| 11 |
-
{"Benchmark": "MATH-500", "Category": "Math", "Pile Dirty": 0.6, "DCLM Dirty": 3.2, "CC202505 Dirty": 0.6, "URL": "https://huggingface.co/datasets/HuggingFaceH4/MATH-500"},
|
| 12 |
-
{"Benchmark": "MGSM", "Category": "Math", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 5.6, "URL": "https://huggingface.co/datasets/juletxara/mgsm"},
|
| 13 |
|
| 14 |
-
{"Benchmark": "HumanEval", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "URL": "https://huggingface.co/datasets/openai/openai_humaneval"},
|
| 15 |
-
{"Benchmark": "HumanEval+", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "URL": "https://huggingface.co/datasets/evalplus/humanevalplus"},
|
| 16 |
-
{"Benchmark": "LiveCodeBench", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "URL": "https://huggingface.co/datasets/livecodebench/code_generation"},
|
| 17 |
-
{"Benchmark": "SWE-bench", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.2, "URL": "https://huggingface.co/datasets/princeton-nlp/SWE-bench_Verified"},
|
| 18 |
-
{"Benchmark": "MBPP", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.4, "CC202505 Dirty": 1.0, "URL": "https://huggingface.co/datasets/google-research-datasets/mbpp"},
|
| 19 |
|
| 20 |
-
{"Benchmark": "ARC-Challenge", "Category": "Commonsense Understanding", "Pile Dirty": 1.8, "DCLM Dirty": 34.1, "CC202505 Dirty": 11.9, "URL": "https://huggingface.co/datasets/allenai/ai2_arc"},
|
| 21 |
-
{"Benchmark": "ARC-Easy", "Category": "Commonsense Understanding", "Pile Dirty": 1.3, "DCLM Dirty": 31.7, "CC202505 Dirty": 5.4, "URL": "https://huggingface.co/datasets/allenai/ai2_arc"},
|
| 22 |
-
{"Benchmark": "CSQA", "Category": "Commonsense Understanding", "Pile Dirty": 0.1, "DCLM Dirty": 1.0, "CC202505 Dirty": 0.1, "URL": "https://huggingface.co/datasets/tau/commonsense_qa"},
|
| 23 |
-
{"Benchmark": "HellaSwag", "Category": "Commonsense Understanding", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "URL": "https://huggingface.co/datasets/Rowan/hellaswag"},
|
| 24 |
-
{"Benchmark": "OpenbookQA", "Category": "Commonsense Understanding", "Pile Dirty": 10.8, "DCLM Dirty": 15.6, "CC202505 Dirty": 14.6, "URL": "https://huggingface.co/datasets/allenai/openbookqa"},
|
| 25 |
-
{"Benchmark": "Social IQa", "Category": "Commonsense Understanding", "Pile Dirty": 0.0, "DCLM Dirty": 0.5, "CC202505 Dirty": 0.2, "URL": "https://huggingface.co/datasets/allenai/social_i_qa"},
|
| 26 |
-
{"Benchmark": "WinoGrande", "Category": "Commonsense Understanding", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "URL": "https://huggingface.co/datasets/allenai/winogrande"},
|
| 27 |
|
| 28 |
-
{"Benchmark": "CoQA", "Category": "Reading Comprehension", "Pile Dirty": 8.0, "DCLM Dirty": 18.4, "CC202505 Dirty": 7.4, "URL": "https://huggingface.co/datasets/stanfordnlp/coqa"},
|
| 29 |
-
{"Benchmark": "SQuAD", "Category": "Reading Comprehension", "Pile Dirty": 2.8, "DCLM Dirty": 40.1, "CC202505 Dirty": 2.7, "URL": "https://huggingface.co/datasets/rajpurkar/squad"}
|
| 30 |
]
|
|
|
|
| 1 |
[
|
| 2 |
+
{"Benchmark": "MMLU", "Category": "Knowledge and Reasoning", "Pile Dirty": 13.2, "DCLM Dirty": 28.4, "CC202505 Dirty": 13.5, "CC202508 Dirty": 9.0, "URL": "https://huggingface.co/datasets/cais/mmlu"},
|
| 3 |
+
{"Benchmark": "MMLU-Pro", "Category": "Knowledge and Reasoning", "Pile Dirty": 5.5, "DCLM Dirty": 16.2, "CC202505 Dirty": 7.1, "CC202508 Dirty": 5.4, "URL": "https://huggingface.co/datasets/TIGER-Lab/MMLU-Pro"},
|
| 4 |
+
{"Benchmark": "BBH", "Category": "Knowledge and Reasoning", "Pile Dirty": 0.0, "DCLM Dirty": 0.1, "CC202505 Dirty": 1.4, "CC202508 Dirty": 1.4, "URL": "https://github.com/suzgunmirac/BIG-Bench-Hard/tree/main/bbh"},
|
| 5 |
+
{"Benchmark": "AGIEval", "Category": "Knowledge and Reasoning", "Pile Dirty": 0.8, "DCLM Dirty": 3.1, "CC202505 Dirty": 2.7, "CC202508 Dirty": 3.6, "URL": "https://github.com/ruixiangcui/AGIEval/tree/main/data/v1_1"},
|
| 6 |
+
{"Benchmark": "GPQA", "Category": "Knowledge and Reasoning", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.89, "CC202508 Dirty": 2.0, "URL": "https://huggingface.co/datasets/Idavidrein/gpqa"},
|
| 7 |
+
{"Benchmark": "HLE", "Category": "Knowledge and Reasoning", "Pile Dirty": 0.0, "DCLM Dirty": 0.3, "CC202505 Dirty": 0.1, "CC202508 Dirty": 0.0, "URL": "https://huggingface.co/datasets/cais/hle"},
|
| 8 |
|
| 9 |
+
{"Benchmark": "AIME_2024", "Category": "Math", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 10.0, "CC202508 Dirty": 3.3, "URL": "https://huggingface.co/datasets/Maxwell-Jia/AIME_2024"},
|
| 10 |
+
{"Benchmark": "GSM8K", "Category": "Math", "Pile Dirty": 0.0, "DCLM Dirty": 0.4, "CC202505 Dirty": 5.0, "CC202508 Dirty": 0.8, "URL": "https://huggingface.co/datasets/openai/gsm8k"},
|
| 11 |
+
{"Benchmark": "MATH-500", "Category": "Math", "Pile Dirty": 0.6, "DCLM Dirty": 3.2, "CC202505 Dirty": 0.6, "CC202508 Dirty": 7.8, "URL": "https://huggingface.co/datasets/HuggingFaceH4/MATH-500"},
|
| 12 |
+
{"Benchmark": "MGSM", "Category": "Math", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 5.6, "CC202508 Dirty": 1.6, "URL": "https://huggingface.co/datasets/juletxara/mgsm"},
|
| 13 |
|
| 14 |
+
{"Benchmark": "HumanEval", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "CC202508 Dirty": 0.6, "URL": "https://huggingface.co/datasets/openai/openai_humaneval"},
|
| 15 |
+
{"Benchmark": "HumanEval+", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "CC202508 Dirty": 0.6, "URL": "https://huggingface.co/datasets/evalplus/humanevalplus"},
|
| 16 |
+
{"Benchmark": "LiveCodeBench", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "CC202508 Dirty": 0.0, "URL": "https://huggingface.co/datasets/livecodebench/code_generation"},
|
| 17 |
+
{"Benchmark": "SWE-bench", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.2, "CC202508 Dirty": 0.2, "URL": "https://huggingface.co/datasets/princeton-nlp/SWE-bench_Verified"},
|
| 18 |
+
{"Benchmark": "MBPP", "Category": "Code", "Pile Dirty": 0.0, "DCLM Dirty": 0.4, "CC202505 Dirty": 1.0, "CC202508 Dirty": 1.4, "URL": "https://huggingface.co/datasets/google-research-datasets/mbpp"},
|
| 19 |
|
| 20 |
+
{"Benchmark": "ARC-Challenge", "Category": "Commonsense Understanding", "Pile Dirty": 1.8, "DCLM Dirty": 34.1, "CC202505 Dirty": 11.9, "CC202508 Dirty": 4.0, "URL": "https://huggingface.co/datasets/allenai/ai2_arc"},
|
| 21 |
+
{"Benchmark": "ARC-Easy", "Category": "Commonsense Understanding", "Pile Dirty": 1.3, "DCLM Dirty": 31.7, "CC202505 Dirty": 5.4, "CC202508 Dirty": 9.5, "URL": "https://huggingface.co/datasets/allenai/ai2_arc"},
|
| 22 |
+
{"Benchmark": "CSQA", "Category": "Commonsense Understanding", "Pile Dirty": 0.1, "DCLM Dirty": 1.0, "CC202505 Dirty": 0.1, "CC202508 Dirty": 0.1, "URL": "https://huggingface.co/datasets/tau/commonsense_qa"},
|
| 23 |
+
{"Benchmark": "HellaSwag", "Category": "Commonsense Understanding", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "CC202508 Dirty": 0.0, "URL": "https://huggingface.co/datasets/Rowan/hellaswag"},
|
| 24 |
+
{"Benchmark": "OpenbookQA", "Category": "Commonsense Understanding", "Pile Dirty": 10.8, "DCLM Dirty": 15.6, "CC202505 Dirty": 14.6, "CC202508 Dirty": 30.2, "URL": "https://huggingface.co/datasets/allenai/openbookqa"},
|
| 25 |
+
{"Benchmark": "Social IQa", "Category": "Commonsense Understanding", "Pile Dirty": 0.0, "DCLM Dirty": 0.5, "CC202505 Dirty": 0.2, "CC202508 Dirty": 4.4, "URL": "https://huggingface.co/datasets/allenai/social_i_qa"},
|
| 26 |
+
{"Benchmark": "WinoGrande", "Category": "Commonsense Understanding", "Pile Dirty": 0.0, "DCLM Dirty": 0.0, "CC202505 Dirty": 0.0, "CC202508 Dirty": 0.0, "URL": "https://huggingface.co/datasets/allenai/winogrande"},
|
| 27 |
|
| 28 |
+
{"Benchmark": "CoQA", "Category": "Reading Comprehension", "Pile Dirty": 8.0, "DCLM Dirty": 18.4, "CC202505 Dirty": 7.4, "CC202508 Dirty": 8.8, "URL": "https://huggingface.co/datasets/stanfordnlp/coqa"},
|
| 29 |
+
{"Benchmark": "SQuAD", "Category": "Reading Comprehension", "Pile Dirty": 2.8, "DCLM Dirty": 40.1, "CC202505 Dirty": 2.7, "CC202508 Dirty": 33.0, "URL": "https://huggingface.co/datasets/rajpurkar/squad"}
|
| 30 |
]
|