duplicated block id: 1
size: 118 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (391:598)
 - pretraining/fairseq/models/pair_bert.py (440:647)

duplicated block id: 2
size: 75 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (470:565)
 - code/run_squad.py (520:615)

duplicated block id: 3
size: 59 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (26:129)
 - pretraining/fairseq/models/pair_bert.py (25:128)

duplicated block id: 4
size: 54 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (367:436)
 - pretraining/fairseq/models/pair_bert.py (258:327)

duplicated block id: 5
size: 54 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (191:261)
 - code/run_squad.py (206:269)

duplicated block id: 6
size: 51 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/adaptive_input.py (16:75)
 - pretraining/fairseq/modules/adaptive_inputs.py (16:75)

duplicated block id: 7
size: 47 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (270:330)
 - pretraining/fairseq/models/pair_bert.py (269:329)

duplicated block id: 8
size: 45 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (378:436)
 - pretraining/fairseq/models/hf_bert.py (270:328)

duplicated block id: 9
size: 43 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (739:784)
 - code/run_squad.py (919:964)

duplicated block id: 10
size: 43 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (156:207)
 - pretraining/fairseq/models/pair_bert.py (155:206)

duplicated block id: 11
size: 37 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (631:677)
 - code/run_squad.py (809:855)

duplicated block id: 12
size: 35 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (402:438)
 - code/run_squad.py (436:472)

duplicated block id: 13
size: 34 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/optimization.py (71:112)
 - pretraining/fairseq/optim/bert_adam.py (94:134)

duplicated block id: 14
size: 33 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (314:361)
 - pretraining/fairseq/models/pair_bert.py (207:254)

duplicated block id: 15
size: 33 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (443:489)
 - pretraining/fairseq/models/hf_bert.py (333:379)

duplicated block id: 16
size: 31 cleaned lines of code
in 2 files:
 - code/run_glue.py (792:828)
 - code/run_tacred.py (384:419)

duplicated block id: 17
size: 30 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (314:357)
 - code/run_squad.py (326:369)

duplicated block id: 18
size: 29 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (231:268)
 - pretraining/fairseq/models/pair_bert.py (230:267)

duplicated block id: 19
size: 29 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (705:737)
 - code/run_squad.py (884:916)

duplicated block id: 20
size: 27 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (785:814)
 - code/run_squad.py (967:996)

duplicated block id: 21
size: 26 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (625:653)
 - pretraining/fairseq/models/pair_bert.py (673:701)

duplicated block id: 22
size: 25 cleaned lines of code
in 2 files:
 - code/run_glue.py (800:828)
 - code/run_mrqa.py (749:776)

duplicated block id: 23
size: 25 cleaned lines of code
in 2 files:
 - code/run_glue.py (800:828)
 - code/run_squad.py (929:956)

duplicated block id: 24
size: 25 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (817:841)
 - code/run_squad.py (999:1023)

duplicated block id: 25
size: 25 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/no_nsp_span_bert_dataset.py (21:45)
 - pretraining/fairseq/data/span_bert_dataset.py (37:62)

duplicated block id: 26
size: 25 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (749:776)
 - code/run_tacred.py (391:419)

duplicated block id: 27
size: 25 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/no_nsp_span_bert_dataset.py (170:198)
 - pretraining/fairseq/data/span_bert_dataset.py (366:394)

duplicated block id: 28
size: 25 cleaned lines of code
in 2 files:
 - code/run_squad.py (929:956)
 - code/run_tacred.py (391:419)

duplicated block id: 29
size: 24 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (378:401)
 - code/run_squad.py (404:427)

duplicated block id: 30
size: 23 cleaned lines of code
in 2 files:
 - code/run_glue.py (349:379)
 - code/run_glue.py (385:415)

duplicated block id: 31
size: 23 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/optimization.py (116:158)
 - pretraining/fairseq/optim/bert_adam.py (136:178)

duplicated block id: 32
size: 22 cleaned lines of code
in 2 files:
 - code/run_glue.py (853:878)
 - code/run_tacred.py (435:460)

duplicated block id: 33
size: 21 cleaned lines of code
in 2 files:
 - code/run_glue.py (981:1002)
 - code/run_tacred.py (549:569)

duplicated block id: 34
size: 21 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (192:221)
 - pretraining/fairseq/models/pair_bert.py (96:125)

duplicated block id: 35
size: 21 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (192:221)
 - pretraining/fairseq/models/hf_bert.py (97:126)

duplicated block id: 36
size: 21 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (853:873)
 - code/run_squad.py (1036:1056)

duplicated block id: 37
size: 21 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (440:463)
 - code/run_squad.py (484:507)

duplicated block id: 38
size: 20 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/no_nsp_span_bert_dataset.py (119:138)
 - pretraining/fairseq/data/span_bert_dataset.py (275:294)

duplicated block id: 39
size: 20 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (159:186)
 - code/run_squad.py (174:201)

duplicated block id: 40
size: 20 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (37:60)
 - pretraining/fairseq/criterions/spanbert_loss.py (35:58)

duplicated block id: 41
size: 19 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (337:361)
 - pretraining/fairseq/models/hf_bert.py (231:255)

duplicated block id: 42
size: 18 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/masking.py (113:131)
 - pretraining/fairseq/data/masking.py (156:174)

duplicated block id: 43
size: 18 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (683:701)
 - code/run_squad.py (862:880)

duplicated block id: 44
size: 18 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (621:641)
 - pretraining/fairseq/models/pair_bert.py (542:562)

duplicated block id: 45
size: 18 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (621:641)
 - pretraining/fairseq/models/hf_bert.py (493:513)

duplicated block id: 46
size: 18 cleaned lines of code
in 2 files:
 - code/run_glue.py (690:711)
 - code/run_tacred.py (304:324)

duplicated block id: 47
size: 17 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/bidirectional_multihead_attention.py (101:124)
 - pretraining/fairseq/modules/multihead_attention.py (313:336)

duplicated block id: 48
size: 17 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (295:312)
 - code/run_squad.py (306:323)

duplicated block id: 49
size: 17 cleaned lines of code
in 2 files:
 - code/run_glue.py (305:329)
 - code/run_glue.py (421:445)

duplicated block id: 50
size: 17 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (289:312)
 - pretraining/fairseq/models/hf_bert.py (184:206)

duplicated block id: 51
size: 17 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (289:312)
 - pretraining/fairseq/models/pair_bert.py (183:205)

duplicated block id: 52
size: 16 cleaned lines of code
in 2 files:
 - code/run_glue.py (103:125)
 - code/run_glue.py (232:254)

duplicated block id: 53
size: 16 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (176:191)
 - pretraining/fairseq/models/hf_bert.py (80:95)

duplicated block id: 54
size: 16 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (133:152)
 - pretraining/fairseq/models/pair_bert.py (132:151)

duplicated block id: 55
size: 16 cleaned lines of code
in 2 files:
 - code/run_glue.py (230:253)
 - code/run_glue.py (305:328)

duplicated block id: 56
size: 16 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (176:191)
 - pretraining/fairseq/models/pair_bert.py (79:94)

duplicated block id: 57
size: 16 cleaned lines of code
in 2 files:
 - code/run_glue.py (230:253)
 - code/run_glue.py (421:444)

duplicated block id: 58
size: 15 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (469:489)
 - pretraining/fairseq/models/pair_bert.py (405:425)

duplicated block id: 59
size: 15 cleaned lines of code
in 2 files:
 - code/run_glue.py (103:124)
 - code/run_glue.py (307:328)

duplicated block id: 60
size: 15 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/masking.py (73:87)
 - pretraining/fairseq/data/masking.py (138:152)

duplicated block id: 61
size: 15 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (279:293)
 - code/run_squad.py (288:302)

duplicated block id: 62
size: 15 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (359:379)
 - pretraining/fairseq/models/pair_bert.py (405:425)

duplicated block id: 63
size: 15 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (587:601)
 - code/run_squad.py (700:714)

duplicated block id: 64
size: 15 cleaned lines of code
in 2 files:
 - code/run_glue.py (103:124)
 - code/run_glue.py (423:444)

duplicated block id: 65
size: 14 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (272:287)
 - pretraining/fairseq/models/hf_bert.py (167:181)

duplicated block id: 66
size: 14 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (707:734)
 - pretraining/fairseq/models/pair_bert.py (628:654)

duplicated block id: 67
size: 14 cleaned lines of code
in 2 files:
 - code/run_glue.py (191:212)
 - code/run_glue.py (305:326)

duplicated block id: 68
size: 14 cleaned lines of code
in 2 files:
 - code/run_glue.py (988:1002)
 - code/run_mrqa.py (927:941)

duplicated block id: 69
size: 14 cleaned lines of code
in 2 files:
 - code/run_glue.py (191:212)
 - code/run_glue.py (421:442)

duplicated block id: 70
size: 14 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (272:287)
 - pretraining/fairseq/models/pair_bert.py (166:180)

duplicated block id: 71
size: 14 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (927:941)
 - code/run_tacred.py (556:569)

duplicated block id: 72
size: 14 cleaned lines of code
in 2 files:
 - code/run_glue.py (191:212)
 - code/run_glue.py (230:251)

duplicated block id: 73
size: 14 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (702:716)
 - pretraining/fairseq/models/pair_bert.py (747:761)

duplicated block id: 74
size: 13 cleaned lines of code
in 2 files:
 - code/run_glue.py (103:122)
 - code/run_glue.py (193:212)

duplicated block id: 75
size: 13 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/bert_loss.py (26:46)
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (28:48)

duplicated block id: 76
size: 13 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (611:623)
 - code/run_squad.py (761:773)

duplicated block id: 77
size: 13 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (70:86)
 - code/run_squad.py (72:88)

duplicated block id: 78
size: 13 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/bert_loss.py (43:58)
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (57:72)

duplicated block id: 79
size: 13 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (891:903)
 - code/run_squad.py (1086:1098)

duplicated block id: 80
size: 13 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (340:356)
 - pretraining/fairseq/models/pair_bert.py (352:368)

duplicated block id: 81
size: 13 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (450:466)
 - pretraining/fairseq/models/pair_bert.py (352:368)

duplicated block id: 82
size: 13 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (119:131)
 - code/run_squad.py (119:131)

duplicated block id: 83
size: 13 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (265:277)
 - code/run_squad.py (271:283)

duplicated block id: 84
size: 13 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/no_nsp_span_bert_dataset.py (229:250)
 - pretraining/fairseq/data/span_bert_dataset.py (428:449)

duplicated block id: 85
size: 12 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (316:335)
 - pretraining/fairseq/models/hf_bert.py (210:229)

duplicated block id: 86
size: 12 cleaned lines of code
in 2 files:
 - code/run_glue.py (988:999)
 - code/run_squad.py (1122:1133)

duplicated block id: 87
size: 12 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (927:938)
 - code/run_squad.py (1122:1133)

duplicated block id: 88
size: 12 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (672:684)
 - pretraining/fairseq/models/pair_bert.py (719:730)

duplicated block id: 89
size: 12 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (57:69)
 - code/run_squad.py (57:69)

duplicated block id: 90
size: 12 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (507:523)
 - pretraining/fairseq/models/hf_bert.py (397:413)

duplicated block id: 91
size: 12 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_loss.py (24:43)
 - pretraining/fairseq/criterions/spanbert_loss.py (27:46)

duplicated block id: 92
size: 12 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (601:612)
 - pretraining/fairseq/models/pair_bert.py (650:661)

duplicated block id: 93
size: 12 cleaned lines of code
in 2 files:
 - code/run_squad.py (1122:1133)
 - code/run_tacred.py (556:567)

duplicated block id: 94
size: 12 cleaned lines of code
in 2 files:
 - code/run_glue.py (368:379)
 - code/run_glue.py (440:451)

duplicated block id: 95
size: 12 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/no_nsp_span_bert_dataset.py (199:219)
 - pretraining/fairseq/data/span_bert_dataset.py (396:416)

duplicated block id: 96
size: 12 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/bidirectional_multihead_attention.py (134:145)
 - pretraining/fairseq/modules/multihead_attention.py (342:353)

duplicated block id: 97
size: 12 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (688:699)
 - pretraining/fairseq/models/pair_bert.py (733:744)

duplicated block id: 98
size: 12 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (914:925)
 - code/run_squad.py (1109:1120)

duplicated block id: 99
size: 12 cleaned lines of code
in 2 files:
 - code/run_glue.py (404:415)
 - code/run_glue.py (440:451)

duplicated block id: 100
size: 12 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (507:523)
 - pretraining/fairseq/models/pair_bert.py (446:462)

duplicated block id: 101
size: 12 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (210:229)
 - pretraining/fairseq/models/pair_bert.py (209:228)

duplicated block id: 102
size: 11 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (82:94)
 - pretraining/fairseq/criterions/spanbert_loss.py (71:83)

duplicated block id: 103
size: 11 cleaned lines of code
in 2 files:
 - code/run_glue.py (867:878)
 - code/run_mrqa.py (803:814)

duplicated block id: 104
size: 11 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (359:370)
 - code/run_squad.py (372:383)

duplicated block id: 105
size: 11 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (88:98)
 - code/run_squad.py (91:101)

duplicated block id: 106
size: 11 cleaned lines of code
in 2 files:
 - pretraining/fairseq/optim/adam.py (78:95)
 - pretraining/fairseq/optim/bert_adam.py (126:142)

duplicated block id: 107
size: 11 cleaned lines of code
in 2 files:
 - code/run_glue.py (867:878)
 - code/run_squad.py (985:996)

duplicated block id: 108
size: 11 cleaned lines of code
in 2 files:
 - code/run_squad.py (985:996)
 - code/run_tacred.py (449:460)

duplicated block id: 109
size: 11 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (658:669)
 - pretraining/fairseq/models/pair_bert.py (706:716)

duplicated block id: 110
size: 11 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (803:814)
 - code/run_tacred.py (449:460)

duplicated block id: 111
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (675:684)
 - pretraining/fairseq/models/pair_bert.py (750:759)

duplicated block id: 112
size: 10 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (832:841)
 - code/run_tacred.py (480:490)

duplicated block id: 113
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (675:684)
 - pretraining/fairseq/models/hf_bert.py (705:714)

duplicated block id: 114
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/bert_loss.py (35:46)
 - pretraining/fairseq/criterions/mlm_loss.py (32:43)

duplicated block id: 115
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_loss.py (32:43)
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (37:48)

duplicated block id: 116
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (705:714)
 - pretraining/fairseq/models/pair_bert.py (721:730)

duplicated block id: 117
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (385:400)
 - code/run_glue.py (421:436)

duplicated block id: 118
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/span_bert_dataset.py (85:94)
 - pretraining/fairseq/data/span_bert_dataset.py (157:166)

duplicated block id: 119
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/bidirectional_multihead_attention.py (83:96)
 - pretraining/fairseq/modules/multihead_attention.py (296:309)

duplicated block id: 120
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (191:206)
 - code/run_glue.py (349:364)

duplicated block id: 121
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (269:284)
 - code/run_glue.py (305:320)

duplicated block id: 122
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (191:206)
 - code/run_glue.py (385:400)

duplicated block id: 123
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (305:320)
 - code/run_glue.py (349:364)

duplicated block id: 124
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (305:320)
 - code/run_glue.py (385:400)

duplicated block id: 125
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (902:912)
 - code/run_mrqa.py (832:841)

duplicated block id: 126
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (230:245)
 - code/run_glue.py (269:284)

duplicated block id: 127
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (269:284)
 - code/run_glue.py (421:436)

duplicated block id: 128
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (349:364)
 - code/run_glue.py (421:436)

duplicated block id: 129
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (191:206)
 - code/run_glue.py (269:284)

duplicated block id: 130
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (269:284)
 - code/run_glue.py (385:400)

duplicated block id: 131
size: 10 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (258:269)
 - pretraining/fairseq/models/pair_bert.py (154:164)

duplicated block id: 132
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (269:284)
 - code/run_glue.py (349:364)

duplicated block id: 133
size: 10 cleaned lines of code
in 2 files:
 - code/run_squad.py (1014:1023)
 - code/run_tacred.py (480:490)

duplicated block id: 134
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/optim/lr_scheduler/fixed_schedule.py (43:56)
 - pretraining/fairseq/optim/lr_scheduler/polynomial_decay_schedule.py (48:61)

duplicated block id: 135
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/bert_loss.py (35:46)
 - pretraining/fairseq/criterions/spanbert_loss.py (35:46)

duplicated block id: 136
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (902:912)
 - code/run_squad.py (1014:1023)

duplicated block id: 137
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/no_nsp_span_bert_dataset.py (67:76)
 - pretraining/fairseq/data/span_bert_dataset.py (130:139)

duplicated block id: 138
size: 10 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/pair_bert.py (721:730)
 - pretraining/fairseq/models/pair_bert.py (750:759)

duplicated block id: 139
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (902:912)
 - code/run_tacred.py (480:490)

duplicated block id: 140
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (230:245)
 - code/run_glue.py (349:364)

duplicated block id: 141
size: 10 cleaned lines of code
in 2 files:
 - code/run_glue.py (230:245)
 - code/run_glue.py (385:400)

duplicated block id: 142
size: 9 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (861:869)
 - code/run_squad.py (870:878)

duplicated block id: 143
size: 9 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (581:591)
 - pretraining/fairseq/models/hf_bert.py (465:475)

duplicated block id: 144
size: 9 cleaned lines of code
in 2 files:
 - code/run_glue.py (103:116)
 - code/run_glue.py (271:284)

duplicated block id: 145
size: 9 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (691:699)
 - code/run_mrqa.py (861:869)

duplicated block id: 146
size: 9 cleaned lines of code
in 2 files:
 - code/run_squad.py (870:878)
 - code/run_squad.py (1044:1052)

duplicated block id: 147
size: 9 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (691:699)
 - code/run_squad.py (1044:1052)

duplicated block id: 148
size: 9 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (581:591)
 - pretraining/fairseq/models/pair_bert.py (514:524)

duplicated block id: 149
size: 9 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/span_bert_dataset.py (119:127)
 - pretraining/fairseq/data/span_bert_dataset.py (187:196)

duplicated block id: 150
size: 9 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (140:148)
 - pretraining/fairseq/models/hf_bert.py (45:53)

duplicated block id: 151
size: 9 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (259:269)
 - pretraining/fairseq/models/hf_bert.py (156:165)

duplicated block id: 152
size: 9 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (571:579)
 - pretraining/fairseq/models/hf_bert.py (455:463)

duplicated block id: 153
size: 9 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/data_utils.py (29:38)
 - pretraining/fairseq/data/data_utils.py (48:57)

duplicated block id: 154
size: 9 cleaned lines of code
in 2 files:
 - code/run_glue.py (103:116)
 - code/run_glue.py (351:364)

duplicated block id: 155
size: 9 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (571:579)
 - pretraining/fairseq/models/pair_bert.py (504:512)

duplicated block id: 156
size: 9 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (96:104)
 - pretraining/fairseq/criterions/spanbert_loss.py (84:92)

duplicated block id: 157
size: 9 cleaned lines of code
in 2 files:
 - code/run_tacred.py (341:349)
 - code/run_tacred.py (502:510)

duplicated block id: 158
size: 9 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (140:148)
 - pretraining/fairseq/models/pair_bert.py (44:52)

duplicated block id: 159
size: 9 cleaned lines of code
in 2 files:
 - code/run_glue.py (103:116)
 - code/run_glue.py (387:400)

duplicated block id: 160
size: 8 cleaned lines of code
in 2 files:
 - code/run_glue.py (56:69)
 - code/run_tacred.py (45:58)

duplicated block id: 161
size: 8 cleaned lines of code
in 2 files:
 - code/mrqa_official_eval.py (54:64)
 - code/run_squad.py (633:642)

duplicated block id: 162
size: 8 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/pair_bert.py (708:715)
 - pretraining/fairseq/models/pair_bert.py (736:743)

duplicated block id: 163
size: 8 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (661:668)
 - pretraining/fairseq/models/pair_bert.py (736:743)

duplicated block id: 164
size: 8 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (691:698)
 - pretraining/fairseq/models/pair_bert.py (708:715)

duplicated block id: 165
size: 8 cleaned lines of code
in 2 files:
 - code/run_glue.py (836:843)
 - code/run_tacred.py (426:433)

duplicated block id: 166
size: 8 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/no_nsp_span_bert_dataset.py (162:169)
 - pretraining/fairseq/data/span_bert_dataset.py (357:364)

duplicated block id: 167
size: 8 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (367:376)
 - pretraining/fairseq/models/hf_bert.py (259:268)

duplicated block id: 168
size: 8 cleaned lines of code
in 2 files:
 - pretraining/fairseq/optim/lr_scheduler/fixed_schedule.py (15:25)
 - pretraining/fairseq/optim/lr_scheduler/polynomial_decay_schedule.py (15:25)

duplicated block id: 169
size: 8 cleaned lines of code
in 2 files:
 - code/run_squad.py (861:868)
 - code/run_squad.py (1035:1042)

duplicated block id: 170
size: 8 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (661:668)
 - pretraining/fairseq/models/hf_bert.py (691:698)

duplicated block id: 171
size: 8 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (707:727)
 - pretraining/fairseq/models/hf_bert.py (579:598)

duplicated block id: 172
size: 8 cleaned lines of code
in 2 files:
 - code/run_glue.py (655:664)
 - code/run_tacred.py (280:288)

duplicated block id: 173
size: 8 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/bert_loss.py (66:75)
 - pretraining/fairseq/criterions/mlm_loss.py (53:62)

duplicated block id: 174
size: 7 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (336:343)
 - pretraining/fairseq/models/pair_bert.py (268:275)

duplicated block id: 175
size: 7 cleaned lines of code
in 2 files:
 - code/run_glue.py (755:762)
 - code/run_squad.py (892:898)

duplicated block id: 176
size: 7 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (119:138)
 - pretraining/fairseq/models/pair_bert.py (25:42)

duplicated block id: 177
size: 7 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (336:343)
 - code/pytorch_pretrained_bert/modeling.py (377:384)

duplicated block id: 178
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/bidirectional_multihead_attention.py (69:78)
 - pretraining/fairseq/modules/multihead_attention.py (285:294)

duplicated block id: 179
size: 7 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/optimization.py (116:122)
 - pretraining/fairseq/optim/adam.py (89:95)

duplicated block id: 180
size: 7 cleaned lines of code
in 2 files:
 - code/run_glue.py (705:711)
 - code/run_squad.py (844:850)

duplicated block id: 181
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/bidirectional_multihead_attention.py (106:114)
 - pretraining/fairseq/modules/multihead_attention.py (176:184)

duplicated block id: 182
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/optim/nag.py (19:31)
 - pretraining/fairseq/optim/sgd.py (19:31)

duplicated block id: 183
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/bert_loss.py (89:95)
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (113:119)

duplicated block id: 184
size: 7 cleaned lines of code
in 2 files:
 - code/run_glue.py (755:762)
 - code/run_mrqa.py (713:720)

duplicated block id: 185
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (771:777)
 - pretraining/fairseq/models/pair_bert.py (816:822)

duplicated block id: 186
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/bidirectional_multihead_attention.py (23:29)
 - pretraining/fairseq/modules/multihead_attention.py (251:257)

duplicated block id: 187
size: 7 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (377:384)
 - pretraining/fairseq/models/pair_bert.py (229:236)

duplicated block id: 188
size: 7 cleaned lines of code
in 2 files:
 - code/run_glue.py (705:711)
 - code/run_mrqa.py (666:672)

duplicated block id: 189
size: 7 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (666:672)
 - code/run_tacred.py (318:324)

duplicated block id: 190
size: 7 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (713:720)
 - code/run_tacred.py (356:363)

duplicated block id: 191
size: 7 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (119:138)
 - pretraining/fairseq/models/hf_bert.py (26:43)

duplicated block id: 192
size: 7 cleaned lines of code
in 2 files:
 - code/run_glue.py (885:891)
 - code/run_tacred.py (466:472)

duplicated block id: 193
size: 7 cleaned lines of code
in 2 files:
 - code/run_squad.py (892:898)
 - code/run_tacred.py (356:363)

duplicated block id: 194
size: 7 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (683:689)
 - code/run_squad.py (1036:1042)

duplicated block id: 195
size: 7 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (683:689)
 - code/run_mrqa.py (853:859)

duplicated block id: 196
size: 7 cleaned lines of code
in 2 files:
 - code/run_glue.py (776:784)
 - code/run_tacred.py (369:377)

duplicated block id: 197
size: 7 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/optimization.py (33:44)
 - pretraining/fairseq/optim/bert_adam.py (58:65)

duplicated block id: 198
size: 7 cleaned lines of code
in 2 files:
 - code/run_glue.py (677:685)
 - code/run_tacred.py (294:302)

duplicated block id: 199
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/multihead_attention.py (176:184)
 - pretraining/fairseq/modules/multihead_attention.py (318:326)

duplicated block id: 200
size: 7 cleaned lines of code
in 2 files:
 - code/run_squad.py (844:850)
 - code/run_tacred.py (318:324)

duplicated block id: 201
size: 7 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (853:859)
 - code/run_squad.py (862:868)

duplicated block id: 202
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (230:237)
 - pretraining/fairseq/models/hf_bert.py (269:276)

duplicated block id: 203
size: 7 cleaned lines of code
in 2 files:
 - code/run_glue.py (755:762)
 - code/run_tacred.py (356:363)

duplicated block id: 204
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/no_nsp_span_bert_dataset.py (220:227)
 - pretraining/fairseq/data/span_bert_dataset.py (418:425)

duplicated block id: 205
size: 7 cleaned lines of code
in 2 files:
 - code/run_glue.py (834:841)
 - code/run_squad.py (961:967)

duplicated block id: 206
size: 7 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/pair_bert.py (229:236)
 - pretraining/fairseq/models/pair_bert.py (268:275)

duplicated block id: 207
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/bert_loss.py (90:95)
 - pretraining/fairseq/criterions/spanbert_loss.py (99:104)

duplicated block id: 208
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_loss.py (73:78)
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (114:119)

duplicated block id: 209
size: 6 cleaned lines of code
in 2 files:
 - code/run_squad.py (922:927)
 - code/run_tacred.py (384:390)

duplicated block id: 210
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (733:738)
 - code/run_tacred.py (340:345)

duplicated block id: 211
size: 6 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (50:55)
 - code/run_squad.py (49:54)

duplicated block id: 212
size: 6 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (802:807)
 - pretraining/fairseq/models/pair_bert.py (828:833)

duplicated block id: 213
size: 6 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/optimization.py (168:181)
 - pretraining/fairseq/optim/bert_adam.py (180:193)

duplicated block id: 214
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (792:799)
 - code/run_squad.py (922:927)

duplicated block id: 215
size: 6 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (378:384)
 - pretraining/fairseq/models/hf_bert.py (231:237)

duplicated block id: 216
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/optim/adam.py (21:27)
 - pretraining/fairseq/optim/bert_adam.py (17:23)

duplicated block id: 217
size: 6 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (742:747)
 - code/run_tacred.py (384:390)

duplicated block id: 218
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_loss.py (73:78)
 - pretraining/fairseq/criterions/spanbert_loss.py (99:104)

duplicated block id: 219
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (29:35)
 - code/run_mrqa.py (28:34)

duplicated block id: 220
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (558:564)
 - code/run_tacred.py (211:217)

duplicated block id: 221
size: 6 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (565:570)
 - code/pytorch_pretrained_bert/tokenization.py (177:182)

duplicated block id: 222
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (792:799)
 - code/run_mrqa.py (742:747)

duplicated block id: 223
size: 6 cleaned lines of code
in 2 files:
 - code/mrqa_official_eval.py (34:42)
 - code/pytorch_pretrained_bert/file_utils.py (103:111)

duplicated block id: 224
size: 6 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (43:48)
 - code/run_squad.py (41:46)

duplicated block id: 225
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (231:237)
 - pretraining/fairseq/models/pair_bert.py (269:275)

duplicated block id: 226
size: 6 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (441:447)
 - pretraining/fairseq/models/pair_bert.py (330:336)

duplicated block id: 227
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (324:329)
 - code/run_glue.py (404:409)

duplicated block id: 228
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/adaptive_inputs.py (27:33)
 - pretraining/fairseq/modules/adaptive_softmax.py (60:66)

duplicated block id: 229
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (967:972)
 - code/run_tacred.py (531:536)

duplicated block id: 230
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/multihead_attention.py (62:67)
 - pretraining/fairseq/modules/multihead_attention.py (268:273)

duplicated block id: 231
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_loss.py (40:46)
 - pretraining/fairseq/criterions/spanbert_loss.py (55:62)

duplicated block id: 232
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/models/hf_bert.py (270:276)
 - pretraining/fairseq/models/pair_bert.py (230:236)

duplicated block id: 233
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (324:329)
 - code/run_glue.py (368:373)

duplicated block id: 234
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (114:119)
 - pretraining/fairseq/criterions/spanbert_loss.py (99:104)

duplicated block id: 235
size: 6 cleaned lines of code
in 2 files:
 - code/run_squad.py (962:967)
 - code/run_tacred.py (426:431)

duplicated block id: 236
size: 6 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (646:651)
 - pretraining/fairseq/models/pair_bert.py (564:569)

duplicated block id: 237
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/bert_loss.py (82:88)
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (105:111)

duplicated block id: 238
size: 6 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (337:343)
 - pretraining/fairseq/models/hf_bert.py (270:276)

duplicated block id: 239
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (288:293)
 - code/run_glue.py (368:373)

duplicated block id: 240
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (169:174)
 - code/run_glue.py (368:373)

duplicated block id: 241
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (288:293)
 - code/run_glue.py (404:409)

duplicated block id: 242
size: 6 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (646:651)
 - pretraining/fairseq/models/hf_bert.py (515:520)

duplicated block id: 243
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (547:552)
 - code/run_tacred.py (204:209)

duplicated block id: 244
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (288:293)
 - code/run_glue.py (440:445)

duplicated block id: 245
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (169:174)
 - code/run_glue.py (404:409)

duplicated block id: 246
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (169:174)
 - code/run_glue.py (440:445)

duplicated block id: 247
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (520:530)
 - code/run_tacred.py (190:195)

duplicated block id: 248
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (288:293)
 - code/run_glue.py (324:329)

duplicated block id: 249
size: 6 cleaned lines of code
in 2 files:
 - code/run_mrqa.py (908:913)
 - code/run_squad.py (1102:1107)

duplicated block id: 250
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/optim/lr_scheduler/fixed_schedule.py (27:33)
 - pretraining/fairseq/optim/lr_scheduler/polynomial_decay_schedule.py (29:35)

duplicated block id: 251
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (926:931)
 - code/run_tacred.py (501:506)

duplicated block id: 252
size: 6 cleaned lines of code
in 2 files:
 - code/pytorch_pretrained_bert/modeling.py (982:988)
 - code/pytorch_pretrained_bert/modeling.py (1117:1123)

duplicated block id: 253
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (747:753)
 - code/run_tacred.py (347:353)

duplicated block id: 254
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/bert_loss.py (90:95)
 - pretraining/fairseq/criterions/mlm_loss.py (73:78)

duplicated block id: 255
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (169:174)
 - code/run_glue.py (288:293)

duplicated block id: 256
size: 6 cleaned lines of code
in 2 files:
 - code/run_glue.py (169:174)
 - code/run_glue.py (324:329)

duplicated block id: 257
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/criterions/mlm_nsp_sbo_loss.py (23:35)
 - pretraining/fairseq/criterions/spanbert_loss.py (22:34)

duplicated block id: 258
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/data/no_nsp_span_bert_dataset.py (53:58)
 - pretraining/fairseq/data/span_bert_dataset.py (65:70)

duplicated block id: 259
size: 6 cleaned lines of code
in 2 files:
 - pretraining/fairseq/modules/adaptive_input.py (27:33)
 - pretraining/fairseq/modules/adaptive_softmax.py (60:66)