Add data from "An Open-Source Data Contamination Report for Large Language Models" (#5)
Browse files- Add data from "An Open-Source Data Contamination Report for Large Language Models" (6169ce28edf11369ba1957b2266cba0bb4da2024)
- Update contamination_report.csv (4831ab13739e53bf664a1b82467a6a2d4794ed7e)
- Fix merge error (2e138568a5af505805315c084367b0c4c8bcab27)
Co-authored-by: Vishaal Udandarao <vishaal27@users.noreply.huggingface.co>
- contamination_report.csv +8 -0
contamination_report.csv
CHANGED
@@ -1,5 +1,13 @@
|
|
1 |
Evaluation Dataset;Subset;Contaminated Source;Model or corpus;Train Split;Development Split;Test Split;Approach;Reference;PR
|
2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
conll2003;;GPT-3.5;model;100.0;100.0;100.0;model-based;https://hitz-zentroa.github.io/lm-contamination/blog/;7
|
4 |
nyu-mll/glue;mnli;GPT-3.5;model;100.0;100.0;;model-based;https://hitz-zentroa.github.io/lm-contamination/blog/;7
|
5 |
rajpurkar/squad_v2;;GPT-3.5;model;100.0;100.0;;model-based;https://hitz-zentroa.github.io/lm-contamination/blog/;7
|
|
|
1 |
Evaluation Dataset;Subset;Contaminated Source;Model or corpus;Train Split;Development Split;Test Split;Approach;Reference;PR
|
2 |
|
3 |
+
|
4 |
+
allenai/ai2_arc;;CommonCrawl;corpus;;;28.7;data-based;https://arxiv.org/abs/2310.17589;5
|
5 |
+
tau/commonsense_qa;;CommonCrawl;corpus;;1.6;;data-based;https://arxiv.org/abs/2310.17589;5
|
6 |
+
winogrande;;CommonCrawl;corpus;;1.1;;data-based;https://arxiv.org/abs/2310.17589;5
|
7 |
+
ceval/ceval-exam;;CommonCrawl;corpus;;45.8;;data-based;https://arxiv.org/abs/2310.17589;5
|
8 |
+
Rowan/hellaswag;;CommonCrawl;corpus;;12.4;;data-based;https://arxiv.org/abs/2310.17589;5
|
9 |
+
cais/mmlu;;CommonCrawl;corpus;;;29.1;data-based;https://arxiv.org/abs/2310.17589;5
|
10 |
+
|
11 |
conll2003;;GPT-3.5;model;100.0;100.0;100.0;model-based;https://hitz-zentroa.github.io/lm-contamination/blog/;7
|
12 |
nyu-mll/glue;mnli;GPT-3.5;model;100.0;100.0;;model-based;https://hitz-zentroa.github.io/lm-contamination/blog/;7
|
13 |
rajpurkar/squad_v2;;GPT-3.5;model;100.0;100.0;;model-based;https://hitz-zentroa.github.io/lm-contamination/blog/;7
|