Added Contamination Info on Old Models: GPT3, FLAN, GLaM, PaLM, PaLM 2 (#13)
Browse files- Update contamination_report.csv (e068ce3ff3a9fc6a23ef66687ce1fb2ffcdc153d)
- Update contamination_report.csv (6febb2668817ca6bab3dee054bc1811a06d4e99a)
- Update contamination_report.csv (0f5ff265a43ef025e95dfd5217592d6f0226494e)
- Fix some format bugs (c714a687cee8b45c649fce9f2d9ef78fe558c144)
- Merge branch 'main' of https://huggingface.co/spaces/CONDA-Workshop/Data-Contamination-Report into pr/13 (b1e5746efc1087663f87cb8c7bcf32b598d959c6)
- Fix some arxiv links (403385af15e9561ffd666af1ba15a8638aa1552e)
Co-authored-by: Ameya Prabhu <AmeyaPrabhu@users.noreply.huggingface.co>
- contamination_report.csv +113 -1
contamination_report.csv
CHANGED
|
@@ -479,9 +479,121 @@ bigbio/mednli;;GPT-3.5;model;0.0;0.0;0.0;model-based;https://arxiv.org/abs/2308.
|
|
| 479 |
|
| 480 |
RadNLI;;GPT-4;model;0.0;0.0;0.0;model-based;https://arxiv.org/abs/2308.08493;8
|
| 481 |
RadNLI;;GPT-3.5;model;0.0;0.0;0.0;model-based;https://arxiv.org/abs/2308.08493;8
|
|
|
|
|
|
|
| 482 |
|
| 483 |
|
| 484 |
openai_humaneval;;EleutherAI/pile;corpus;;;12.2;data-based;https://arxiv.org/abs/2403.04811;12
|
| 485 |
mbpp;;EleutherAI/pile;corpus;;;3.6;data-based;https://arxiv.org/abs/2403.04811;12
|
| 486 |
openai_humaneval;;bigcode/the-stack;corpus;;;18.9;data-based;https://arxiv.org/abs/2403.04811;12
|
| 487 |
-
mbpp;;bigcode/the-stack;corpus;;;20.8;data-based;https://arxiv.org/abs/2403.04811;12
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 479 |
|
| 480 |
RadNLI;;GPT-4;model;0.0;0.0;0.0;model-based;https://arxiv.org/abs/2308.08493;8
|
| 481 |
RadNLI;;GPT-3.5;model;0.0;0.0;0.0;model-based;https://arxiv.org/abs/2308.08493;8
|
| 482 |
+
RadNLI;;GPT-4;model;0.0;0.0;0.0;model-based;https://arxiv.org/abs/2308.08493;8
|
| 483 |
+
RadNLI;;GPT-3.5;model;0.0;0.0;0.0;model-based;https://arxiv.org/abs/2308.08493;8
|
| 484 |
|
| 485 |
|
| 486 |
openai_humaneval;;EleutherAI/pile;corpus;;;12.2;data-based;https://arxiv.org/abs/2403.04811;12
|
| 487 |
mbpp;;EleutherAI/pile;corpus;;;3.6;data-based;https://arxiv.org/abs/2403.04811;12
|
| 488 |
openai_humaneval;;bigcode/the-stack;corpus;;;18.9;data-based;https://arxiv.org/abs/2403.04811;12
|
| 489 |
+
mbpp;;bigcode/the-stack;corpus;;;20.8;data-based;https://arxiv.org/abs/2403.04811;12
|
| 490 |
+
|
| 491 |
+
quac;;GPT-3;model;;99.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 492 |
+
rajpurkar/squad_v2;;GPT-3;model;;94.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 493 |
+
ucinlp/drop;;GPT-3;model;;93.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 494 |
+
Symbol Insertion;;GPT-3;model;;86.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 495 |
+
stanfordnlp/coqa;;GPT-3;model;;64.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 496 |
+
super_glue;record;GPT-3;model;;61.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 497 |
+
winograd_wsc;;GPT-3;model;;;60.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 498 |
+
super_glue;boolq;GPT-3;model;;60.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 499 |
+
super_glue;multirc;GPT-3;model;;59.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 500 |
+
race;high;GPT-3;model;;;45.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 501 |
+
cimec/lambada;;GPT-3;model;;;43.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 502 |
+
super_glue;wsc;GPT-3;model;;40.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 503 |
+
piqa;;GPT-3;model;;29.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 504 |
+
wmt/wmt16;en-de;GPT-3;model;;;25.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 505 |
+
wmt/wmt16;de-en;GPT-3;model;;;25.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 506 |
+
race;middle;GPT-3;model;;;25.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 507 |
+
rmanluo/RoG-webqsp;;GPT-3;model;;;21.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 508 |
+
wmt/wmt16;en-ro;GPT-3;model;;;21.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 509 |
+
wmt/wmt16;ro-en;GPT-3;model;;;21.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 510 |
+
facebook/anli;test_r1;GPT-3;model;;;20.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 511 |
+
facebook/anli;test_r2;GPT-3;model;;;18.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 512 |
+
mandarjoshi/trivia_qa;;GPT-3;model;;17.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 513 |
+
facebook/anli;test_r3;GPT-3;model;;;16.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 514 |
+
wmt/wmt16;fr-en;GPT-3;model;;;14.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 515 |
+
wmt/wmt16;en-fr;GPT-3;model;;;14.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 516 |
+
super_glue;rte;GPT-3;model;;8.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 517 |
+
super_glue;wic;GPT-3;model;;8.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 518 |
+
super_glue;cb;GPT-3;model;;7.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 519 |
+
Reversed Words;;GPT-3;model;;7.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 520 |
+
Anagrams 2;;GPT-3;model;;7.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 521 |
+
allenai/openbookqa;;GPT-3;model;;;6.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 522 |
+
ibragim-bad/arc_easy;;GPT-3;model;;;4.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 523 |
+
Anagrams 1;;GPT-3;model;;3.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 524 |
+
ibragim-bad/arc_challenge;;GPT-3;model;;;3.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 525 |
+
super_glue;copa;GPT-3;model;;3.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 526 |
+
Rowan/hellaswag;;GPT-3;model;;2.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 527 |
+
natural_questions;;GPT-3;model;;;1.0;data-based;https://arxiv.org/abs/2005.14165;13
|
| 528 |
+
Cycled Letters;;GPT-3;model;;1.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 529 |
+
SAT Analogies;;GPT-3;model;;1.0;;data-based;https://arxiv.org/abs/2005.14165;13
|
| 530 |
+
|
| 531 |
+
EdinburghNLP/xsum;;PaLM 2;model;;;42.0;data-based;https://arxiv.org/abs/2305.10403;13
|
| 532 |
+
csebuetnlp/xlsum;;PaLM 2;model;;;46.9;data-based;https://arxiv.org/abs/2305.10403;13
|
| 533 |
+
wiki_lingua;;PaLM 2;model;;;9.0;data-based;https://arxiv.org/abs/2305.10403;13
|
| 534 |
+
|
| 535 |
+
winograd_wsc;;PaLM;model;;;38.5;data-based;https://arxiv.org/abs/2204.02311;13
|
| 536 |
+
rmanluo/RoG-webqsp;;PaLM;model;;;26.7;data-based;https://arxiv.org/abs/2204.02311;13
|
| 537 |
+
super_glue;wsc;PaLM;model;;;36.8;data-based;https://arxiv.org/abs/2204.02311;13
|
| 538 |
+
mandarjoshi/trivia_qa;;PaLM;model;;19.9;;data-based;https://arxiv.org/abs/2204.02311;13
|
| 539 |
+
rajpurkar/squad_v2;;PaLM;model;;85.2;;data-based;https://arxiv.org/abs/2204.02311;13
|
| 540 |
+
super_glue;record;PaLM;model;;43.4;;data-based;https://arxiv.org/abs/2204.02311;13
|
| 541 |
+
cimec/lambada;;PaLM;model;;;29.3;data-based;https://arxiv.org/abs/2204.02311;13
|
| 542 |
+
super_glue;cb;PaLM;model;;48.2;;data-based;https://arxiv.org/abs/2204.02311;13
|
| 543 |
+
ibragim-bad/arc_easy;;PaLM;model;;;30.4;data-based;https://arxiv.org/abs/2204.02311;13
|
| 544 |
+
ibragim-bad/arc_challenge;;PaLM;model;;;24.7;data-based;https://arxiv.org/abs/2204.02311;13
|
| 545 |
+
|
| 546 |
+
winograd_wsc;;GLaM;model;;67.3;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 547 |
+
winogrande;;GLaM;model;;;0.3;data-based;https://arxiv.org/abs/2112.06905;13
|
| 548 |
+
super_glue;wic;GLaM;model;;8.2;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 549 |
+
super_glue;wsc;GLaM;model;;57.5;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 550 |
+
mandarjoshi/trivia_qa;;GLaM;model;;18.8;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 551 |
+
story_cloze;;GLaM;model;;100.0;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 552 |
+
rajpurkar/squad_v2;;GLaM;model;;94.6;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 553 |
+
super_glue;record;GLaM;model;;98.6;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 554 |
+
super_glue;rte;GLaM;model;;54.9;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 555 |
+
race;middle;GLaM;model;;58.4;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 556 |
+
race;high;GLaM;model;;74.0;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 557 |
+
quac;;GLaM;model;;99.9;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 558 |
+
piqa;;GLaM;model;;49.8;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 559 |
+
allenai/openbookqa;;GLaM;model;;20.0;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 560 |
+
natural_questions;;GLaM;model;;3.9;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 561 |
+
super_glue;multirc;GLaM;model;;68.8;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 562 |
+
cimec/lambada;;GLaM;model;;;21.8;data-based;https://arxiv.org/abs/2112.06905;13
|
| 563 |
+
Rowan/hellaswag;;GLaM;model;;19.8;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 564 |
+
stanfordnlp/coqa;;GLaM;model;;;75.0;data-based;https://arxiv.org/abs/2112.06905;13
|
| 565 |
+
super_glue;copa;GLaM;model;;3.0;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 566 |
+
super_glue;cb;GLaM;model;;26.8;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 567 |
+
super_glue;boolq;GLaM;model;;92.1;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 568 |
+
ibragim-bad/arc_easy;;GLaM;model;;32.5;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 569 |
+
ibragim-bad/arc_challenge;;GLaM;model;;31.8;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 570 |
+
facebook/anli;dev_r3;GLaM;model;;40.7;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 571 |
+
facebook/anli;dev_r2;GLaM;model;;96.8;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 572 |
+
facebook/anli;dev_r1;GLaM;model;;96.2;;data-based;https://arxiv.org/abs/2112.06905;13
|
| 573 |
+
|
| 574 |
+
winogrande;;FLAN;model;;;0.2;data-based;https://arxiv.org/abs/2109.01652;13
|
| 575 |
+
mandarjoshi/trivia_qa;;FLAN;model;;22.8;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 576 |
+
story_cloze;;FLAN;model;;0.4;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 577 |
+
rajpurkar/squad_v2;;FLAN;model;;99.1;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 578 |
+
wmt/wmt16;ro-en;FLAN;model;;;12.4;data-based;https://arxiv.org/abs/2109.01652;13
|
| 579 |
+
super_glue;record;FLAN;model;;68.0;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 580 |
+
super_glue;rte;FLAN;model;;33.9;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 581 |
+
piqa;;FLAN;model;;51.3;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 582 |
+
allenai/openbookqa;;FLAN;model;;15.0;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 583 |
+
natural_questions;;FLAN;model;;3.2;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 584 |
+
super_glue;multirc;FLAN;model;;59.3;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 585 |
+
Rowan/hellaswag;;FLAN;model;;34.5;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 586 |
+
wmt/wmt16;fr-en;FLAN;model;;;25.3;data-based;https://arxiv.org/abs/2109.01652;13
|
| 587 |
+
wmt/wmt16;en-ro;FLAN;model;;;12.4;data-based;https://arxiv.org/abs/2109.01652;13
|
| 588 |
+
wmt/wmt16;en-fr;FLAN;model;;;25.3;data-based;https://arxiv.org/abs/2109.01652;13
|
| 589 |
+
wmt/wmt16;en-de;FLAN;model;;;14.3;data-based;https://arxiv.org/abs/2109.01652;13
|
| 590 |
+
wmt/wmt16;de-en;FLAN;model;;;14.3;data-based;https://arxiv.org/abs/2109.01652;13
|
| 591 |
+
ucinlp/drop;;FLAN;model;;99.4;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 592 |
+
super_glue;copa;FLAN;model;;9.0;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 593 |
+
super_glue;cb;FLAN;model;;5.4;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 594 |
+
super_glue;boolq;FLAN;model;;23.1;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 595 |
+
ibragim-bad/arc_easy;;FLAN;model;;20.2;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 596 |
+
ibragim-bad/arc_challenge;;FLAN;model;;15.6;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 597 |
+
facebook/anli;dev_r3;FLAN;model;;40.2;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 598 |
+
facebook/anli;dev_r2;FLAN;model;;97.9;;data-based;https://arxiv.org/abs/2109.01652;13
|
| 599 |
+
facebook/anli;dev_r1;FLAN;model;;98.6;;data-based;https://arxiv.org/abs/2109.01652;13
|