Upload tokenizer

None.txt

@@ -0,0 +1,1000 @@
+<pad>	0
+<unk>	0
+<s>	0
+</s>	0
+s	-2.98325
+▁	-3.33613
+▁the	-3.53659
+,	-3.96136
+▁of	-4.11741
+e	-4.13254
+.	-4.1679
+▁a	-4.32755
+▁and	-4.33125
+▁in	-4.37771
+ed	-4.40148
+d	-4.43306
+t	-4.4655
+y	-4.53252
+ing	-4.61545
+er	-4.6614
+a	-4.69442
+l	-4.74331
+▁to	-4.74857
+n	-4.76571
+o	-4.80687
+▁,	-4.92362
+▁is	-4.93935
+r	-4.97962
+al	-4.99362
+in	-5.09839
+ar	-5.10752
+p	-5.16879
+f	-5.17396
+or	-5.20309
+c	-5.21379
+g	-5.21537
+▁The	-5.26093
+re	-5.33666
+-	-5.37802
+b	-5.39736
+ly	-5.41386
+▁.	-5.4411
+en	-5.45186
+i	-5.47261
+▁(	-5.50639
+▁f	-5.51352
+an	-5.54095
+▁as	-5.54212
+ur	-5.55306
+▁S	-5.57071
+▁re	-5.58107
+es	-5.58272
+▁for	-5.62901
+ic	-5.63297
+m	-5.64497
+u	-5.65596
+▁A	-5.65812
+le	-5.66668
+it	-5.68914
+▁by	-5.70114
+▁was	-5.75532
+on	-5.79453
+ri	-5.79803
+▁B	-5.80188
+st	-5.81431
+'	-5.82833
+▁with	-5.83968
+th	-5.84353
+▁on	-5.87698
+h	-5.88915
+▁P	-5.89336
+w	-5.89774
+▁or	-5.91013
+ra	-5.92151
+ation	-5.92539
+▁be	-5.9376
+il	-5.98519
+ent	-5.9883
+▁an	-5.99257
+▁19	-6.01172
+ion	-6.01493
+at	-6.02207
+▁"	-6.02294
+▁c	-6.02546
+ch	-6.02773
+)	-6.03923
+k	-6.04155
+▁co	-6.04329
+▁are	-6.05606
+ro	-6.0652
+▁C	-6.07878
+▁p	-6.11511
+v	-6.1246
+us	-6.13632
+▁de	-6.14746
+▁st	-6.15681
+▁that	-6.16069
+1	-6.16222
+se	-6.17232
+ce	-6.18052
+ir	-6.18303
+▁from	-6.18936
+5	-6.19203
+is	-6.22921
+"	-6.23156
+▁D	-6.25522
+el	-6.25566
+ne	-6.27447
+te	-6.28033
+ad	-6.29411
+▁G	-6.29495
+ment	-6.29817
+am	-6.29912
+▁F	-6.29947
+x	-6.30442
+▁T	-6.32441
+li	-6.32652
+▁m	-6.33549
+ol	-6.34257
+ive	-6.35499
+9	-6.3607
+un	-6.38235
+▁con	-6.38422
+id	-6.3859
+ate	-6.38848
+est	-6.3947
+▁In	-6.39701
+ul	-6.40068
+ut	-6.41284
+0	-6.41971
+▁t	-6.42085
+▁H	-6.43357
+et	-6.44432
+2	-6.44628
+as	-6.44759
+▁se	-6.4536
+6	-6.45453
+▁b	-6.46467
+7	-6.46587
+la	-6.4908
+8	-6.49474
+ist	-6.49804
+▁ex	-6.50137
+▁g	-6.50505
+S	-6.50922
+▁h	-6.51419
+4	-6.52466
+▁which	-6.54796
+▁at	-6.58108
+▁W	-6.5819
+3	-6.58465
+im	-6.5961
+om	-6.59781
+A	-6.60127
+age	-6.60365
+ies	-6.60467
+▁w	-6.62078
+ver	-6.62924
+▁N	-6.63071
+ers	-6.63181
+▁E	-6.63282
+C	-6.63669
+ve	-6.63671
+▁e	-6.65324
+▁has	-6.66208
+▁it	-6.67689
+ter	-6.68174
+ity	-6.68823
+mp	-6.70881
+▁his	-6.72489
+▁d	-6.73116
+um	-6.7349
+ge	-6.73673
+▁M	-6.74175
+me	-6.74328
+▁r	-6.74662
+▁1	-6.75046
+qu	-6.75189
+▁I	-6.75752
+▁United	-6.76143
+nd	-6.76592
+lo	-6.76616
+▁ch	-6.77466
+▁also	-6.7776
+▁J	-6.79458
+▁po	-6.80132
+),	-6.80651
+▁It	-6.81648
+▁pro	-6.81762
+and	-6.81885
+▁he	-6.8234
+▁have	-6.83154
+z	-6.8364
+▁O	-6.84215
+ill	-6.84536
+ant	-6.84949
+▁R	-6.85995
+▁what	-6.88048
+ated	-6.88381
+▁States	-6.88593
+ine	-6.89957
+▁di	-6.90808
+ight	-6.90919
+ance	-6.91298
+ph	-6.9194
+▁other	-6.92562
+;	-6.93669
+co	-6.94291
+▁U	-6.94648
+tic	-6.95711
+ide	-6.96324
+ta	-6.96777
+▁its	-6.9678
+ect	-6.97397
+ud	-6.97493
+iv	-6.98038
+▁who	-6.98122
+▁mo	-6.98793
+I	-6.99739
+ot	-7.01328
+▁not	-7.01789
+▁most	-7.02148
+▁v	-7.02259
+ci	-7.02331
+▁were	-7.03457
+tion	-7.04212
+ow	-7.04501
+ry	-7.04542
+▁2	-7.05202
+op	-7.05275
+▁first	-7.0554
+▁K	-7.05685
+all	-7.0605
+min	-7.06175
+▁can	-7.06893
+ia	-7.07865
+ac	-7.08502
+▁sp	-7.09015
+▁pa	-7.09396
+▁one	-7.09743
+).	-7.10594
+ther	-7.11186
+pe	-7.1143
+av	-7.12624
+mo	-7.12761
+▁Ch	-7.13342
+ical	-7.13516
+pp	-7.13969
+▁American	-7.14112
+▁un	-7.14512
+ian	-7.14524
+ig	-7.14606
+▁been	-7.15507
+ence	-7.15667
+▁used	-7.15929
+▁so	-7.16047
+ap	-7.17222
+▁St	-7.17543
+ven	-7.18062
+ke	-7.18374
+old	-7.20342
+▁ma	-7.2115
+▁form	-7.21604
+land	-7.21658
+▁ca	-7.21738
+▁state	-7.21891
+ck	-7.22935
+hi	-7.23329
+ted	-7.24357
+▁la	-7.25031
+▁such	-7.2514
+▁pre	-7.25425
+port	-7.2548
+ie	-7.25505
+D	-7.25589
+▁their	-7.26005
+▁part	-7.26338
+able	-7.26411
+▁Mo	-7.2669
+em	-7.26987
+T	-7.27079
+ma	-7.27164
+ub	-7.27451
+▁He	-7.27531
+ial	-7.27582
+ag	-7.27696
+ction	-7.27945
+P	-7.29408
+up	-7.31156
+▁V	-7.31241
+ard	-7.31839
+ous	-7.32009
+▁me	-7.32062
+ell	-7.32153
+:	-7.32453
+▁but	-7.32773
+he	-7.33481
+▁no	-7.34059
+▁over	-7.34226
+▁many	-7.34398
+ha	-7.35318
+mb	-7.36081
+bo	-7.36412
+if	-7.36603
+les	-7.36806
+ten	-7.37112
+▁into	-7.3734
+ally	-7.37849
+ust	-7.37887
+ish	-7.37962
+▁known	-7.38086
+▁two	-7.38213
+ary	-7.38731
+ang	-7.39591
+▁sh	-7.39651
+ho	-7.40195
+▁Ma	-7.4021
+▁all	-7.40284
+ti	-7.40569
+▁ho	-7.40924
+▁L	-7.41168
+ab	-7.41502
+com	-7.42251
+▁name	-7.4232
+▁3	-7.42434
+uc	-7.42528
+ak	-7.42852
+to	-7.43329
+j	-7.43401
+day	-7.44705
+▁j	-7.45545
+B	-7.45711
+▁Re	-7.4583
+–	-7.46056
+▁use	-7.46287
+cent	-7.46858
+▁more	-7.47239
+▁18	-7.47506
+mi	-7.47913
+▁Co	-7.48482
+ture	-7.49362
+po	-7.50349
+ca	-7.50676
+▁when	-7.50816
+▁ba	-7.51262
+E	-7.51342
+und	-7.51527
+▁br	-7.51879
+200	-7.52528
+na	-7.53136
+ach	-7.53575
+▁act	-7.53834
+one	-7.53944
+▁time	-7.54097
+▁play	-7.54584
+ric	-7.55652
+M	-7.55855
+ue	-7.55971
+▁may	-7.56174
+ress	-7.56233
+tra	-7.56266
+▁include	-7.56512
+▁high	-7.56513
+▁car	-7.5676
+▁film	-7.56876
+ative	-7.56917
+▁La	-7.57192
+▁term	-7.57839
+no	-7.58717
+▁some	-7.59031
+lin	-7.59239
+▁New	-7.59409
+low	-7.59432
+▁do	-7.595
+▁after	-7.59602
+tro	-7.59984
+ru	-7.60526
+▁than	-7.6062
+▁bo	-7.61657
+ok	-7.61903
+▁number	-7.62558
+▁world	-7.62936
+ron	-7.63126
+ition	-7.63362
+ack	-7.63495
+ug	-7.63745
+vo	-7.64372
+▁per	-7.64738
+▁where	-7.65822
+▁lo	-7.66202
+man	-7.66506
+▁had	-7.6653
+▁Ro	-7.66706
+R	-7.6693
+N	-7.67234
+▁this	-7.67473
+▁De	-7.68762
+ank	-7.69171
+▁War	-7.69455
+▁inter	-7.70337
+▁Mar	-7.70572
+act	-7.70587
+▁k	-7.71029
+the	-7.71248
+▁gr	-7.71417
+mon	-7.71636
+▁her	-7.72386
+ual	-7.72539
+▁An	-7.73173
+▁only	-7.7342
+L	-7.7428
+▁system	-7.74584
+▁acc	-7.75146
+ship	-7.75202
+ton	-7.75294
+ities	-7.75685
+▁out	-7.75767
+▁Ar	-7.76284
+round	-7.76685
+▁This	-7.76688
+ism	-7.76781
+▁series	-7.77408
+▁refer	-7.77481
+io	-7.77484
+ice	-7.77969
+00	-7.78432
+▁work	-7.78498
+ip	-7.78704
+ized	-7.78924
+▁million	-7.79162
+F	-7.79665
+cul	-7.79869
+over	-7.79984
+▁years	-7.80147
+▁war	-7.80264
+ep	-7.80304
+O	-7.80382
+▁Lo	-7.8093
+during	-7.80962
+▁4	-7.81569
+▁how	-7.82293
+di	-7.82509
+▁common	-7.83726
+▁dis	-7.8391
+▁including	-7.83956
+▁game	-7.84703
+ick	-7.85306
+▁World	-7.8537
+▁under	-7.8545
+▁between	-7.86084
+ration	-7.86217
+ev	-7.87153
+▁well	-7.87304
+▁Ho	-7.8788
+way	-7.87917
+ize	-7.88827
+▁trans	-7.89243
+came	-7.89553
+present	-7.8977
+ga	-7.90352
+imp	-7.90424
+▁20	-7.91065
+▁large	-7.91385
+▁new	-7.92043
+▁record	-7.92244
+▁four	-7.9228
+▁we	-7.92322
+▁major	-7.93266
+H	-7.93297
+▁often	-7.93609
+▁group	-7.94299
+uch	-7.94534
+▁up	-7.94593
+▁develop	-7.94817
+▁produce	-7.95324
+▁people	-7.96394
+ner	-7.96649
+%	-7.97192
+tter	-7.97777
+▁For	-7.97864
+▁plan	-7.98045
+▁Bo	-7.98192
+▁called	-7.98275
+bu	-7.99349
+▁about	-7.99356
+▁Me	-7.99467
+▁population	-7.99625
+con	-7.99954
+▁end	-8.00591
+▁second	-8.01283
+▁mi	-8.01704
+▁area	-8.02148
+▁show	-8.02283
+/	-8.02399
+▁Con	-8.0241
+ttle	-8.03169
+▁since	-8.03253
+▁three	-8.03533
+▁To	-8.03668
+▁vi	-8.03818
+V	-8.0431
+ward	-8.04836
+▁season	-8.05255
+ib	-8.05421
+▁both	-8.0575
+▁America	-8.06135
+▁year	-8.0631
+?	-8.07004
+▁South	-8.073
+▁city	-8.07512
+▁through	-8.07597
+gu	-8.07666
+cri	-8.07767
+▁team	-8.08249
+▁product	-8.08424
+▁found	-8.09697
+▁main	-8.10388
+▁law	-8.10683
+▁sub	-8.1072
+▁Th	-8.10747
+va	-8.11156
+▁character	-8.11215
+▁follow	-8.12453
+▁government	-8.12763
+▁Mi	-8.12877
+▁North	-8.13132
+▁made	-8.13142
+▁man	-8.13487
+ign	-8.13576
+▁creat	-8.14272
+east	-8.1504
+▁human	-8.15615
+▁song	-8.15634
+ator	-8.15765
+▁Le	-8.16108
+▁lead	-8.17645
+▁different	-8.17884
+while	-8.17924
+direct	-8.18304
+,000	-8.18884
+▁2010	-8.19207
+▁long	-8.19886
+▁right	-8.2005
+▁John	-8.2021
+▁power	-8.2024
+▁does	-8.20394
+X	-8.20547
+▁several	-8.2089
+▁largest	-8.21363
+▁countries	-8.21565
+▁general	-8.21566
+▁life	-8.21571
+ible	-8.22191
+▁each	-8.22319
+G	-8.2286
+▁own	-8.23473
+▁US	-8.23481
+▁history	-8.23978
+▁On	-8.24717
+▁17	-8.26111
+▁person	-8.26485
+▁single	-8.26853
+▁mean	-8.26999
+form	-8.27906
+▁2011	-8.28257
+▁member	-8.2826
+▁album	-8.28988
+▁national	-8.28992
+▁Com	-8.29855
+$	-8.30094
+▁public	-8.30466
+▁bi	-8.30766
+▁leg	-8.30841
+▁result	-8.30842
+▁later	-8.30898
+▁Award	-8.30927
+▁President	-8.31214
+▁process	-8.31217
+▁usually	-8.3159
+▁water	-8.3162
+—	-8.31967
+▁century	-8.31968
+▁appear	-8.31968
+▁music	-8.32404
+▁success	-8.32726
+▁service	-8.33108
+▁type	-8.33118
+▁president	-8.33491
+▁perform	-8.34315
+60	-8.35939
+▁Li	-8.36016
+ness	-8.3606
+▁list	-8.36139
+▁featur	-8.36218
+▁operat	-8.36292
+U	-8.36662
+born	-8.36936
+▁released	-8.37568
+▁original	-8.37602
+▁Col	-8.37608
+▁before	-8.37806
+▁small	-8.37808
+▁cause	-8.38501
+▁cell	-8.38604
+▁early	-8.38633
+▁event	-8.38638
+ium	-8.38872
+▁design	-8.39437
+▁Man	-8.3989
+▁language	-8.4024
+▁would	-8.40651
+▁National	-8.40663
+mmuni	-8.41256
+log	-8.41337
+▁serve	-8.4141
+▁current	-8.42735
+▁country	-8.43238
+▁provide	-8.44008
+▁place	-8.4419
+vision	-8.44221
+ology	-8.4486
+▁German	-8.44866
+▁She	-8.44989
+▁electro	-8.45295
+▁live	-8.45679
+Y	-8.45722
+▁disc	-8.45957
+▁popular	-8.46161
+▁will	-8.46457
+▁death	-8.46631
+▁English	-8.4881
+▁same	-8.48844
+▁great	-8.49289
+▁based	-8.49458
+▁won	-8.49533
+tract	-8.49707
+▁2012	-8.50162
+▁(19	-8.50623
+▁sign	-8.50888
+▁role	-8.50985
+▁British	-8.51073
+▁10	-8.51193
+▁program	-8.51531
+▁region	-8.52019
+▁unti	-8.52285
+W	-8.53129
+hose	-8.53731
+▁Africa	-8.53858
+▁take	-8.54016
+view	-8.5449
+ified	-8.55788
+struct	-8.55972
+spi	-8.56224
+▁occur	-8.57207
+▁considered	-8.57694
+▁period	-8.57696
+▁book	-8.57841
+▁slave	-8.58197
+▁short	-8.58758
+writ	-8.59109
+time	-8.59155
+▁television	-8.59172
+▁energy	-8.5967
+▁contain	-8.59676
+▁now	-8.59796
+▁require	-8.60172
+▁health	-8.60182
+▁body	-8.60206
+▁start	-8.60416
+▁began	-8.60709
+▁near	-8.61413
+▁count	-8.61699
+▁official	-8.62195
+▁home	-8.62675
+▁level	-8.62709
+▁Act	-8.62953
+▁economic	-8.63223
+▁political	-8.63223
+▁example	-8.63224
+▁After	-8.63239
+▁2009	-8.63243
+▁become	-8.63249
+▁data	-8.6328
+▁support	-8.63742
+▁Europe	-8.63864
+▁because	-8.64262
+▁effect	-8.64262
+▁standard	-8.64264
+▁south	-8.64309
+▁European	-8.65189
+▁class	-8.6532
+▁various	-8.65841
+▁York	-8.65841
+▁company	-8.65841
+▁modern	-8.65842
+▁total	-8.65873
+▁back	-8.66028
+ization	-8.66893
+▁tax	-8.66965
+▁among	-8.67466
+▁continu	-8.67987
+▁remain	-8.67997
+▁Union	-8.67998
+▁Earth	-8.68
+▁market	-8.6853
+▁multi	-8.69089
+K	-8.69409
+é	-8.69626
+▁third	-8.69627
+▁control	-8.6963
+▁north	-8.69726
+▁computer	-8.70178
+mission	-8.70196
+▁Republic	-8.70734
+▁military	-8.71292
+▁version	-8.7186
+▁six	-8.71863
+ological	-8.71869
+spect	-8.72371
+▁However	-8.72419
+▁individual	-8.72419
+▁device	-8.72421
+▁Japan	-8.72421
+▁federal	-8.72422
+▁like	-8.72453
+▁word	-8.73044
+▁make	-8.73128
+▁allow	-8.73458
+▁located	-8.73561
+Z	-8.74133
+▁Great	-8.74153
+▁open	-8.74226
+▁Roman	-8.74916
+▁office	-8.74921
+▁Canada	-8.75293
+▁East	-8.75446
+▁though	-8.75472
+▁100	-8.75545
+ootball	-8.75878
+▁2008	-8.76485
+rmal	-8.76922
+▁establish	-8.77653
+▁addition	-8.78253
+field	-8.78262
+▁estimate	-8.78858
+▁species	-8.7946
+▁civil	-8.79464
+▁video	-8.79465
+▁issue	-8.80662
+▁January	-8.80683
+▁against	-8.80684
+ograph	-8.80684
+▁along	-8.80688
+▁title	-8.80721
+▁special	-8.81301
+▁engine	-8.81301
+▁colon	-8.81368
+▁specific	-8.81922
+▁tradition	-8.82547
+▁international	-8.82547
+▁qua	-8.83622
+▁April	-8.83809
+▁Some	-8.83892
+▁school	-8.84446
+▁measure	-8.84447
+▁free	-8.84828
+▁League	-8.85087
+▁family	-8.85087
+▁worldwide	-8.85087
+▁element	-8.85097
+▁although	-8.85732
+▁House	-8.85756
+▁food	-8.86383
+▁particular	-8.87035
+▁Kingdom	-8.87035
+▁Congress	-8.87035
+▁social	-8.87701
+▁July	-8.8772
+&	-8.88355
+▁California	-8.88355
+▁June	-8.88382
+▁range	-8.88441
+cycl	-8.88772
+▁sometime	-8.89091
+uff	-8.89115
+▁consist	-8.89121
+▁held	-8.8913
+▁describe	-8.91049
+▁written	-8.91052
+▁source	-8.91054
+▁space	-8.91088
+▁origin	-8.91407
+▁network	-8.91734
+▁author	-8.91735
+ability	-8.91736
+▁function	-8.91737
+▁treat	-8.91873
+▁India	-8.92403
+▁November	-8.92423
+▁increase	-8.92423
+▁artist	-8.92428
+▁2007	-8.92446
+▁culture	-8.93126
+▁Super	-8.93138
+▁Christian	-8.93817
+▁super	-8.94533
+▁associat	-8.95231
+▁blood	-8.95945
+town	-8.96151
+▁natural	-8.97389
+▁France	-8.97399
+dependent	-8.98119
+▁vitamin	-8.98119
+▁animal	-8.9814
+▁model	-8.98146
+▁rock	-8.98175
+▁Australia	-8.98854
+▁September	-8.98854
+▁business	-8.98854
+▁significant	-8.99595
+▁University	-9.00341
+▁organization	-9.00341
+▁education	-9.00342
+▁throughout	-9.00344
+▁compet	-9.00347
+ification	-9.011
+▁story	-9.01217
+▁International	-9.0185
+▁important	-9.0185
+▁research	-9.01851
+▁2006	-9.01876
+▁career	-9.01964
+▁October	-9.03383
+▁August	-9.03384
+▁women	-9.03391
+▁Although	-9.04158
+▁initial	-9.04197
+▁full	-9.04731
+▁billion	-9.04943
+▁novel	-9.04949
+▁limit	-9.0573
+▁Washington	-9.0652
+▁Bowl	-9.06552
+▁global	-9.07321
+▁widely	-9.07326
+▁another	-9.07342
+▁England	-9.08127
+▁disease	-9.08128
+▁With	-9.08167
+▁media	-9.0849
+▁surface	-9.0894
+▁oppos	-9.08941
+▁Qu	-9.09002
+▁collect	-9.09766
+▁Since	-9.09805
+▁help	-9.10654
+▁release	-9.11081
+▁close	-9.11479
+rough	-9.1175
+▁similar	-9.1226
+▁French	-9.1226
+▁white	-9.12262
+▁election	-9.12265
+▁information	-9.12268
+▁employ	-9.13107
+▁influence	-9.13107
+▁capital	-9.13107
+▁regard	-9.13122
+▁children	-9.13962
+▁episode	-9.13962
+▁Greek	-9.13962
+▁return	-9.13964
+▁1990	-9.13968
+▁What	-9.13984
+▁making	-9.1399
+▁given	-9.13997
+J	-9.1477
+▁February	-9.14824
+▁amount	-9.14824
+▁psych	-9.14827
+▁application	-9.15694
+▁replac	-9.15695
+▁value	-9.15699
+▁every	-9.15723
+produc	-9.15995
+▁December	-9.16571
+▁typically	-9.16571
+▁movement	-9.16573
+▁left	-9.16578
+▁however	-9.1658
+▁previous	-9.17456
+▁claim	-9.17462
+▁Windows	-9.18349
+▁observ	-9.18349
+▁material	-9.18349
+▁object	-9.19249
+▁professional	-9.19249
+wood	-9.19263
+▁local	-9.19306
+▁trade	-9.1939
+▁Internet	-9.20159
+ought	-9.20434
+▁especial	-9.21076
+▁Western	-9.21076
+▁either	-9.21083
+pproximate	-9.22002
+▁County	-9.22002
+▁primary	-9.22936
+▁property	-9.22936
+▁account	-9.22937
+▁separate	-9.22938
+▁factor	-9.22948
+▁down	-9.22964
+▁commercial	-9.2388
+▁structure	-9.24835
+▁economy	-9.26765
+▁pressure	-9.26765
+▁attack	-9.26776
+▁condition	-9.26778
+▁decade	-9.26783
+▁week	-9.26791
+▁Confederate	-9.27745
+▁complete	-9.27745
+▁Spanish	-9.28735
+▁across	-9.28736
+▁growth	-9.28736
+▁central	-9.28738
+▁available	-9.29735
+▁certain	-9.29735
+▁2013	-9.29736
+▁You	-9.29745
+▁Whil	-9.30017
+▁primari	-9.30745
+▁contribut	-9.31766
+▁Party	-9.31768
+▁drug	-9.31777
+▁Democrat	-9.32797
+▁campaign	-9.32797
+▁subsequent	-9.32797
+▁focus	-9.32797
+▁elected	-9.32798
+craft	-9.32802
+NFL	-9.32802
+▁environment	-9.33838
+▁chemical	-9.33839
+▁entire	-9.33845
+▁radio	-9.34074
+▁William	-9.34891
+▁complex	-9.34891
+▁financial	-9.35955
+▁George	-9.35955
+▁league	-9.35955
+▁policy	-9.35956
+▁project	-9.35956
+▁compos	-9.35958
+▁litera	-9.36277
+▁physical	-9.3703
+▁historical	-9.3703
+▁hydro	-9.37031
+▁defeat	-9.37041
+▁regulat	-9.37057
+▁According	-9.38117
+▁method	-9.38117
+▁generat	-9.38195
+known	-9.39007
+▁celebrat	-9.39216
+▁built	-9.39216
+▁practice	-9.40327
+▁purpose	-9.41451
+▁credit	-9.41452
+▁affect	-9.42587
+▁degree	-9.4259
+▁beginning	-9.43736
+▁symbol	-9.43736
+▁science	-9.43738
+▁Central	-9.449
+▁industry	-9.46076
+▁private	-9.46076
+▁subject	-9.46076
+▁western	-9.46076
+▁district	-9.47266
+▁community	-9.47266
+▁problem	-9.47266
+▁determine	-9.47268
+▁Constitution	-9.48471
+▁Texas	-9.48471
+▁distinct	-9.48476
+▁except	-9.48476
+▁Association	-9.4969
+▁institution	-9.4969
+▁nuclear	-9.4969
+▁technology	-9.4969
+▁Empire	-9.49691
+▁regular	-9.49691
+▁concept	-9.49694
+▁Mexico	-9.50925
+▁announced	-9.50925
+Q	-9.69917
+q	-9.90537
+(	-9.98015

sentencepiece_tokenizer.py

@@ -0,0 +1,82 @@
+from transformers import PreTrainedTokenizer
+import sentencepiece as spm
+import os
+from logging import getLogger
+from typing import List
+
+logger = getLogger()
+
+class SPTokenizer(PreTrainedTokenizer):
+    """Tokenizing and encoding/decoding text using SentencePiece."""
+
+    def __init__(self, model_path: str, vocab_file: str, **kwargs):
+        # Add 'vocab_file' to kwargs if it's not present
+        if 'vocab_file' not in kwargs:
+            kwargs['vocab_file'] = vocab_file
+
+        super().__init__(**kwargs)
+
+        # Store initialization arguments
+        self.init_kwargs = {"model_path": model_path, "vocab_file": vocab_file, **kwargs}
+
+        # Reload tokenizer
+        assert os.path.isfile(model_path), model_path
+        self.sp_model = spm.SentencePieceProcessor(model_file=model_path)
+        logger.info(f"Reloaded SentencePiece model from {model_path}")
+
+        # BOS / EOS token IDs
+        self.n_words: int = self.sp_model.vocab_size()
+        self.bos_id: int = self.sp_model.bos_id()
+        self.eos_id: int = self.sp_model.eos_id()
+        self.pad_id: int = self.sp_model.pad_id()
+        logger.info(
+            f"#words: {self.n_words} - BOS ID: {self.bos_id} - EOS ID: {self.eos_id}"
+        )
+        assert self.sp_model.vocab_size() == self.sp_model.get_piece_size()
+
+    def encode(self, s: str, bos: bool, eos: bool) -> List[int]:
+        t = self.sp_model.encode(s)
+        if bos:
+            t = [self.bos_id] + t
+        if eos:
+            t = t + [self.eos_id]
+        return t
+
+    def decode(self, t: List[int]) -> str:
+        return self.sp_model.decode(t)
+    @property
+    def vocab_size(self):
+        """Size of the base vocabulary (without the added tokens)."""
+        return self.n_words
+
+    def save_pretrained(self, save_directory: str, max_shard_size=None, safe_serialization=None):
+        # Save the SentencePiece model file
+        super().save_pretrained(save_directory, max_shard_size=None, safe_serialization=None)
+
+
+        # Save the vocabulary to a file
+        self.save_vocabulary(save_directory, filename_prefix="vocab")
+
+    @classmethod
+    def from_pretrained(cls, pretrained_model_name_or_path: str, **kwargs):
+        # Load the SentencePiece model file
+        model_file = os.path.join(pretrained_model_name_or_path, "spmodel_wikiqa.model")
+
+        # Load the vocabulary file path
+        vocab_file = os.path.join(pretrained_model_name_or_path, "spmodel_wikiqa.vocab")
+
+        return cls(model_path=model_file, vocab_file=vocab_file, **kwargs)
+        #return super().from_pretrained(cls, pretrained_model_name_or_path, **kwargs)
+
+    def get_vocab(self):
+        # Read vocabulary from the file
+        with open(self.init_kwargs["vocab_file"], "r", encoding="utf-8") as f:
+            vocab = {word.strip(): i for i, word in enumerate(f)}
+        return vocab
+
+    def save_vocabulary(self, save_directory, filename_prefix):
+        vocab_file = os.path.join(save_directory, f"{filename_prefix}.txt")
+        with open(vocab_file, "w", encoding="utf-8") as f:
+            for word, index in sorted(self.get_vocab().items(), key=lambda x: x[1]):
+                f.write(f"{word}\n")
+        return (vocab_file,)

special_tokens_map.json

@@ -0,0 +1 @@
+{}

tokenizer_config.json

@@ -0,0 +1,14 @@
+{
+  "added_tokens_decoder": {},
+  "auto_map": {
+    "AutoTokenizer": [
+      "sentencepiece_tokenizer.Tokenizer",
+      null
+    ]
+  },
+  "clean_up_tokenization_spaces": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "model_path": "./spmodel_wikiqa.model",
+  "tokenizer_class": "Tokenizer",
+  "vocab_file": "./spmodel_wikiqa.vocab"
+}

vocab.txt

@@ -0,0 +1,1000 @@
+<pad>	0
+<unk>	0
+<s>	0
+</s>	0
+s	-2.98325
+▁	-3.33613
+▁the	-3.53659
+,	-3.96136
+▁of	-4.11741
+e	-4.13254
+.	-4.1679
+▁a	-4.32755
+▁and	-4.33125
+▁in	-4.37771
+ed	-4.40148
+d	-4.43306
+t	-4.4655
+y	-4.53252
+ing	-4.61545
+er	-4.6614
+a	-4.69442
+l	-4.74331
+▁to	-4.74857
+n	-4.76571
+o	-4.80687
+▁,	-4.92362
+▁is	-4.93935
+r	-4.97962
+al	-4.99362
+in	-5.09839
+ar	-5.10752
+p	-5.16879
+f	-5.17396
+or	-5.20309
+c	-5.21379
+g	-5.21537
+▁The	-5.26093
+re	-5.33666
+-	-5.37802
+b	-5.39736
+ly	-5.41386
+▁.	-5.4411
+en	-5.45186
+i	-5.47261
+▁(	-5.50639
+▁f	-5.51352
+an	-5.54095
+▁as	-5.54212
+ur	-5.55306
+▁S	-5.57071
+▁re	-5.58107
+es	-5.58272
+▁for	-5.62901
+ic	-5.63297
+m	-5.64497
+u	-5.65596
+▁A	-5.65812
+le	-5.66668
+it	-5.68914
+▁by	-5.70114
+▁was	-5.75532
+on	-5.79453
+ri	-5.79803
+▁B	-5.80188
+st	-5.81431
+'	-5.82833
+▁with	-5.83968
+th	-5.84353
+▁on	-5.87698
+h	-5.88915
+▁P	-5.89336
+w	-5.89774
+▁or	-5.91013
+ra	-5.92151
+ation	-5.92539
+▁be	-5.9376
+il	-5.98519
+ent	-5.9883
+▁an	-5.99257
+▁19	-6.01172
+ion	-6.01493
+at	-6.02207
+▁"	-6.02294
+▁c	-6.02546
+ch	-6.02773
+)	-6.03923
+k	-6.04155
+▁co	-6.04329
+▁are	-6.05606
+ro	-6.0652
+▁C	-6.07878
+▁p	-6.11511
+v	-6.1246
+us	-6.13632
+▁de	-6.14746
+▁st	-6.15681
+▁that	-6.16069
+1	-6.16222
+se	-6.17232
+ce	-6.18052
+ir	-6.18303
+▁from	-6.18936
+5	-6.19203
+is	-6.22921
+"	-6.23156
+▁D	-6.25522
+el	-6.25566
+ne	-6.27447
+te	-6.28033
+ad	-6.29411
+▁G	-6.29495
+ment	-6.29817
+am	-6.29912
+▁F	-6.29947
+x	-6.30442
+▁T	-6.32441
+li	-6.32652
+▁m	-6.33549
+ol	-6.34257
+ive	-6.35499
+9	-6.3607
+un	-6.38235
+▁con	-6.38422
+id	-6.3859
+ate	-6.38848
+est	-6.3947
+▁In	-6.39701
+ul	-6.40068
+ut	-6.41284
+0	-6.41971
+▁t	-6.42085
+▁H	-6.43357
+et	-6.44432
+2	-6.44628
+as	-6.44759
+▁se	-6.4536
+6	-6.45453
+▁b	-6.46467
+7	-6.46587
+la	-6.4908
+8	-6.49474
+ist	-6.49804
+▁ex	-6.50137
+▁g	-6.50505
+S	-6.50922
+▁h	-6.51419
+4	-6.52466
+▁which	-6.54796
+▁at	-6.58108
+▁W	-6.5819
+3	-6.58465
+im	-6.5961
+om	-6.59781
+A	-6.60127
+age	-6.60365
+ies	-6.60467
+▁w	-6.62078
+ver	-6.62924
+▁N	-6.63071
+ers	-6.63181
+▁E	-6.63282
+C	-6.63669
+ve	-6.63671
+▁e	-6.65324
+▁has	-6.66208
+▁it	-6.67689
+ter	-6.68174
+ity	-6.68823
+mp	-6.70881
+▁his	-6.72489
+▁d	-6.73116
+um	-6.7349
+ge	-6.73673
+▁M	-6.74175
+me	-6.74328
+▁r	-6.74662
+▁1	-6.75046
+qu	-6.75189
+▁I	-6.75752
+▁United	-6.76143
+nd	-6.76592
+lo	-6.76616
+▁ch	-6.77466
+▁also	-6.7776
+▁J	-6.79458
+▁po	-6.80132
+),	-6.80651
+▁It	-6.81648
+▁pro	-6.81762
+and	-6.81885
+▁he	-6.8234
+▁have	-6.83154
+z	-6.8364
+▁O	-6.84215
+ill	-6.84536
+ant	-6.84949
+▁R	-6.85995
+▁what	-6.88048
+ated	-6.88381
+▁States	-6.88593
+ine	-6.89957
+▁di	-6.90808
+ight	-6.90919
+ance	-6.91298
+ph	-6.9194
+▁other	-6.92562
+;	-6.93669
+co	-6.94291
+▁U	-6.94648
+tic	-6.95711
+ide	-6.96324
+ta	-6.96777
+▁its	-6.9678
+ect	-6.97397
+ud	-6.97493
+iv	-6.98038
+▁who	-6.98122
+▁mo	-6.98793
+I	-6.99739
+ot	-7.01328
+▁not	-7.01789
+▁most	-7.02148
+▁v	-7.02259
+ci	-7.02331
+▁were	-7.03457
+tion	-7.04212
+ow	-7.04501
+ry	-7.04542
+▁2	-7.05202
+op	-7.05275
+▁first	-7.0554
+▁K	-7.05685
+all	-7.0605
+min	-7.06175
+▁can	-7.06893
+ia	-7.07865
+ac	-7.08502
+▁sp	-7.09015
+▁pa	-7.09396
+▁one	-7.09743
+).	-7.10594
+ther	-7.11186
+pe	-7.1143
+av	-7.12624
+mo	-7.12761
+▁Ch	-7.13342
+ical	-7.13516
+pp	-7.13969
+▁American	-7.14112
+▁un	-7.14512
+ian	-7.14524
+ig	-7.14606
+▁been	-7.15507
+ence	-7.15667
+▁used	-7.15929
+▁so	-7.16047
+ap	-7.17222
+▁St	-7.17543
+ven	-7.18062
+ke	-7.18374
+old	-7.20342
+▁ma	-7.2115
+▁form	-7.21604
+land	-7.21658
+▁ca	-7.21738
+▁state	-7.21891
+ck	-7.22935
+hi	-7.23329
+ted	-7.24357
+▁la	-7.25031
+▁such	-7.2514
+▁pre	-7.25425
+port	-7.2548
+ie	-7.25505
+D	-7.25589
+▁their	-7.26005
+▁part	-7.26338
+able	-7.26411
+▁Mo	-7.2669
+em	-7.26987
+T	-7.27079
+ma	-7.27164
+ub	-7.27451
+▁He	-7.27531
+ial	-7.27582
+ag	-7.27696
+ction	-7.27945
+P	-7.29408
+up	-7.31156
+▁V	-7.31241
+ard	-7.31839
+ous	-7.32009
+▁me	-7.32062
+ell	-7.32153
+:	-7.32453
+▁but	-7.32773
+he	-7.33481
+▁no	-7.34059
+▁over	-7.34226
+▁many	-7.34398
+ha	-7.35318
+mb	-7.36081
+bo	-7.36412
+if	-7.36603
+les	-7.36806
+ten	-7.37112
+▁into	-7.3734
+ally	-7.37849
+ust	-7.37887
+ish	-7.37962
+▁known	-7.38086
+▁two	-7.38213
+ary	-7.38731
+ang	-7.39591
+▁sh	-7.39651
+ho	-7.40195
+▁Ma	-7.4021
+▁all	-7.40284
+ti	-7.40569
+▁ho	-7.40924
+▁L	-7.41168
+ab	-7.41502
+com	-7.42251
+▁name	-7.4232
+▁3	-7.42434
+uc	-7.42528
+ak	-7.42852
+to	-7.43329
+j	-7.43401
+day	-7.44705
+▁j	-7.45545
+B	-7.45711
+▁Re	-7.4583
+–	-7.46056
+▁use	-7.46287
+cent	-7.46858
+▁more	-7.47239
+▁18	-7.47506
+mi	-7.47913
+▁Co	-7.48482
+ture	-7.49362
+po	-7.50349
+ca	-7.50676
+▁when	-7.50816
+▁ba	-7.51262
+E	-7.51342
+und	-7.51527
+▁br	-7.51879
+200	-7.52528
+na	-7.53136
+ach	-7.53575
+▁act	-7.53834
+one	-7.53944
+▁time	-7.54097
+▁play	-7.54584
+ric	-7.55652
+M	-7.55855
+ue	-7.55971
+▁may	-7.56174
+ress	-7.56233
+tra	-7.56266
+▁include	-7.56512
+▁high	-7.56513
+▁car	-7.5676
+▁film	-7.56876
+ative	-7.56917
+▁La	-7.57192
+▁term	-7.57839
+no	-7.58717
+▁some	-7.59031
+lin	-7.59239
+▁New	-7.59409
+low	-7.59432
+▁do	-7.595
+▁after	-7.59602
+tro	-7.59984
+ru	-7.60526
+▁than	-7.6062
+▁bo	-7.61657
+ok	-7.61903
+▁number	-7.62558
+▁world	-7.62936
+ron	-7.63126
+ition	-7.63362
+ack	-7.63495
+ug	-7.63745
+vo	-7.64372
+▁per	-7.64738
+▁where	-7.65822
+▁lo	-7.66202
+man	-7.66506
+▁had	-7.6653
+▁Ro	-7.66706
+R	-7.6693
+N	-7.67234
+▁this	-7.67473
+▁De	-7.68762
+ank	-7.69171
+▁War	-7.69455
+▁inter	-7.70337
+▁Mar	-7.70572
+act	-7.70587
+▁k	-7.71029
+the	-7.71248
+▁gr	-7.71417
+mon	-7.71636
+▁her	-7.72386
+ual	-7.72539
+▁An	-7.73173
+▁only	-7.7342
+L	-7.7428
+▁system	-7.74584
+▁acc	-7.75146
+ship	-7.75202
+ton	-7.75294
+ities	-7.75685
+▁out	-7.75767
+▁Ar	-7.76284
+round	-7.76685
+▁This	-7.76688
+ism	-7.76781
+▁series	-7.77408
+▁refer	-7.77481
+io	-7.77484
+ice	-7.77969
+00	-7.78432
+▁work	-7.78498
+ip	-7.78704
+ized	-7.78924
+▁million	-7.79162
+F	-7.79665
+cul	-7.79869
+over	-7.79984
+▁years	-7.80147
+▁war	-7.80264
+ep	-7.80304
+O	-7.80382
+▁Lo	-7.8093
+during	-7.80962
+▁4	-7.81569
+▁how	-7.82293
+di	-7.82509
+▁common	-7.83726
+▁dis	-7.8391
+▁including	-7.83956
+▁game	-7.84703
+ick	-7.85306
+▁World	-7.8537
+▁under	-7.8545
+▁between	-7.86084
+ration	-7.86217
+ev	-7.87153
+▁well	-7.87304
+▁Ho	-7.8788
+way	-7.87917
+ize	-7.88827
+▁trans	-7.89243
+came	-7.89553
+present	-7.8977
+ga	-7.90352
+imp	-7.90424
+▁20	-7.91065
+▁large	-7.91385
+▁new	-7.92043
+▁record	-7.92244
+▁four	-7.9228
+▁we	-7.92322
+▁major	-7.93266
+H	-7.93297
+▁often	-7.93609
+▁group	-7.94299
+uch	-7.94534
+▁up	-7.94593
+▁develop	-7.94817
+▁produce	-7.95324
+▁people	-7.96394
+ner	-7.96649
+%	-7.97192
+tter	-7.97777
+▁For	-7.97864
+▁plan	-7.98045
+▁Bo	-7.98192
+▁called	-7.98275
+bu	-7.99349
+▁about	-7.99356
+▁Me	-7.99467
+▁population	-7.99625
+con	-7.99954
+▁end	-8.00591
+▁second	-8.01283
+▁mi	-8.01704
+▁area	-8.02148
+▁show	-8.02283
+/	-8.02399
+▁Con	-8.0241
+ttle	-8.03169
+▁since	-8.03253
+▁three	-8.03533
+▁To	-8.03668
+▁vi	-8.03818
+V	-8.0431
+ward	-8.04836
+▁season	-8.05255
+ib	-8.05421
+▁both	-8.0575
+▁America	-8.06135
+▁year	-8.0631
+?	-8.07004
+▁South	-8.073
+▁city	-8.07512
+▁through	-8.07597
+gu	-8.07666
+cri	-8.07767
+▁team	-8.08249
+▁product	-8.08424
+▁found	-8.09697
+▁main	-8.10388
+▁law	-8.10683
+▁sub	-8.1072
+▁Th	-8.10747
+va	-8.11156
+▁character	-8.11215
+▁follow	-8.12453
+▁government	-8.12763
+▁Mi	-8.12877
+▁North	-8.13132
+▁made	-8.13142
+▁man	-8.13487
+ign	-8.13576
+▁creat	-8.14272
+east	-8.1504
+▁human	-8.15615
+▁song	-8.15634
+ator	-8.15765
+▁Le	-8.16108
+▁lead	-8.17645
+▁different	-8.17884
+while	-8.17924
+direct	-8.18304
+,000	-8.18884
+▁2010	-8.19207
+▁long	-8.19886
+▁right	-8.2005
+▁John	-8.2021
+▁power	-8.2024
+▁does	-8.20394
+X	-8.20547
+▁several	-8.2089
+▁largest	-8.21363
+▁countries	-8.21565
+▁general	-8.21566
+▁life	-8.21571
+ible	-8.22191
+▁each	-8.22319
+G	-8.2286
+▁own	-8.23473
+▁US	-8.23481
+▁history	-8.23978
+▁On	-8.24717
+▁17	-8.26111
+▁person	-8.26485
+▁single	-8.26853
+▁mean	-8.26999
+form	-8.27906
+▁2011	-8.28257
+▁member	-8.2826
+▁album	-8.28988
+▁national	-8.28992
+▁Com	-8.29855
+$	-8.30094
+▁public	-8.30466
+▁bi	-8.30766
+▁leg	-8.30841
+▁result	-8.30842
+▁later	-8.30898
+▁Award	-8.30927
+▁President	-8.31214
+▁process	-8.31217
+▁usually	-8.3159
+▁water	-8.3162
+—	-8.31967
+▁century	-8.31968
+▁appear	-8.31968
+▁music	-8.32404
+▁success	-8.32726
+▁service	-8.33108
+▁type	-8.33118
+▁president	-8.33491
+▁perform	-8.34315
+60	-8.35939
+▁Li	-8.36016
+ness	-8.3606
+▁list	-8.36139
+▁featur	-8.36218
+▁operat	-8.36292
+U	-8.36662
+born	-8.36936
+▁released	-8.37568
+▁original	-8.37602
+▁Col	-8.37608
+▁before	-8.37806
+▁small	-8.37808
+▁cause	-8.38501
+▁cell	-8.38604
+▁early	-8.38633
+▁event	-8.38638
+ium	-8.38872
+▁design	-8.39437
+▁Man	-8.3989
+▁language	-8.4024
+▁would	-8.40651
+▁National	-8.40663
+mmuni	-8.41256
+log	-8.41337
+▁serve	-8.4141
+▁current	-8.42735
+▁country	-8.43238
+▁provide	-8.44008
+▁place	-8.4419
+vision	-8.44221
+ology	-8.4486
+▁German	-8.44866
+▁She	-8.44989
+▁electro	-8.45295
+▁live	-8.45679
+Y	-8.45722
+▁disc	-8.45957
+▁popular	-8.46161
+▁will	-8.46457
+▁death	-8.46631
+▁English	-8.4881
+▁same	-8.48844
+▁great	-8.49289
+▁based	-8.49458
+▁won	-8.49533
+tract	-8.49707
+▁2012	-8.50162
+▁(19	-8.50623
+▁sign	-8.50888
+▁role	-8.50985
+▁British	-8.51073
+▁10	-8.51193
+▁program	-8.51531
+▁region	-8.52019
+▁unti	-8.52285
+W	-8.53129
+hose	-8.53731
+▁Africa	-8.53858
+▁take	-8.54016
+view	-8.5449
+ified	-8.55788
+struct	-8.55972
+spi	-8.56224
+▁occur	-8.57207
+▁considered	-8.57694
+▁period	-8.57696
+▁book	-8.57841
+▁slave	-8.58197
+▁short	-8.58758
+writ	-8.59109
+time	-8.59155
+▁television	-8.59172
+▁energy	-8.5967
+▁contain	-8.59676
+▁now	-8.59796
+▁require	-8.60172
+▁health	-8.60182
+▁body	-8.60206
+▁start	-8.60416
+▁began	-8.60709
+▁near	-8.61413
+▁count	-8.61699
+▁official	-8.62195
+▁home	-8.62675
+▁level	-8.62709
+▁Act	-8.62953
+▁economic	-8.63223
+▁political	-8.63223
+▁example	-8.63224
+▁After	-8.63239
+▁2009	-8.63243
+▁become	-8.63249
+▁data	-8.6328
+▁support	-8.63742
+▁Europe	-8.63864
+▁because	-8.64262
+▁effect	-8.64262
+▁standard	-8.64264
+▁south	-8.64309
+▁European	-8.65189
+▁class	-8.6532
+▁various	-8.65841
+▁York	-8.65841
+▁company	-8.65841
+▁modern	-8.65842
+▁total	-8.65873
+▁back	-8.66028
+ization	-8.66893
+▁tax	-8.66965
+▁among	-8.67466
+▁continu	-8.67987
+▁remain	-8.67997
+▁Union	-8.67998
+▁Earth	-8.68
+▁market	-8.6853
+▁multi	-8.69089
+K	-8.69409
+é	-8.69626
+▁third	-8.69627
+▁control	-8.6963
+▁north	-8.69726
+▁computer	-8.70178
+mission	-8.70196
+▁Republic	-8.70734
+▁military	-8.71292
+▁version	-8.7186
+▁six	-8.71863
+ological	-8.71869
+spect	-8.72371
+▁However	-8.72419
+▁individual	-8.72419
+▁device	-8.72421
+▁Japan	-8.72421
+▁federal	-8.72422
+▁like	-8.72453
+▁word	-8.73044
+▁make	-8.73128
+▁allow	-8.73458
+▁located	-8.73561
+Z	-8.74133
+▁Great	-8.74153
+▁open	-8.74226
+▁Roman	-8.74916
+▁office	-8.74921
+▁Canada	-8.75293
+▁East	-8.75446
+▁though	-8.75472
+▁100	-8.75545
+ootball	-8.75878
+▁2008	-8.76485
+rmal	-8.76922
+▁establish	-8.77653
+▁addition	-8.78253
+field	-8.78262
+▁estimate	-8.78858
+▁species	-8.7946
+▁civil	-8.79464
+▁video	-8.79465
+▁issue	-8.80662
+▁January	-8.80683
+▁against	-8.80684
+ograph	-8.80684
+▁along	-8.80688
+▁title	-8.80721
+▁special	-8.81301
+▁engine	-8.81301
+▁colon	-8.81368
+▁specific	-8.81922
+▁tradition	-8.82547
+▁international	-8.82547
+▁qua	-8.83622
+▁April	-8.83809
+▁Some	-8.83892
+▁school	-8.84446
+▁measure	-8.84447
+▁free	-8.84828
+▁League	-8.85087
+▁family	-8.85087
+▁worldwide	-8.85087
+▁element	-8.85097
+▁although	-8.85732
+▁House	-8.85756
+▁food	-8.86383
+▁particular	-8.87035
+▁Kingdom	-8.87035
+▁Congress	-8.87035
+▁social	-8.87701
+▁July	-8.8772
+&	-8.88355
+▁California	-8.88355
+▁June	-8.88382
+▁range	-8.88441
+cycl	-8.88772
+▁sometime	-8.89091
+uff	-8.89115
+▁consist	-8.89121
+▁held	-8.8913
+▁describe	-8.91049
+▁written	-8.91052
+▁source	-8.91054
+▁space	-8.91088
+▁origin	-8.91407
+▁network	-8.91734
+▁author	-8.91735
+ability	-8.91736
+▁function	-8.91737
+▁treat	-8.91873
+▁India	-8.92403
+▁November	-8.92423
+▁increase	-8.92423
+▁artist	-8.92428
+▁2007	-8.92446
+▁culture	-8.93126
+▁Super	-8.93138
+▁Christian	-8.93817
+▁super	-8.94533
+▁associat	-8.95231
+▁blood	-8.95945
+town	-8.96151
+▁natural	-8.97389
+▁France	-8.97399
+dependent	-8.98119
+▁vitamin	-8.98119
+▁animal	-8.9814
+▁model	-8.98146
+▁rock	-8.98175
+▁Australia	-8.98854
+▁September	-8.98854
+▁business	-8.98854
+▁significant	-8.99595
+▁University	-9.00341
+▁organization	-9.00341
+▁education	-9.00342
+▁throughout	-9.00344
+▁compet	-9.00347
+ification	-9.011
+▁story	-9.01217
+▁International	-9.0185
+▁important	-9.0185
+▁research	-9.01851
+▁2006	-9.01876
+▁career	-9.01964
+▁October	-9.03383
+▁August	-9.03384
+▁women	-9.03391
+▁Although	-9.04158
+▁initial	-9.04197
+▁full	-9.04731
+▁billion	-9.04943
+▁novel	-9.04949
+▁limit	-9.0573
+▁Washington	-9.0652
+▁Bowl	-9.06552
+▁global	-9.07321
+▁widely	-9.07326
+▁another	-9.07342
+▁England	-9.08127
+▁disease	-9.08128
+▁With	-9.08167
+▁media	-9.0849
+▁surface	-9.0894
+▁oppos	-9.08941
+▁Qu	-9.09002
+▁collect	-9.09766
+▁Since	-9.09805
+▁help	-9.10654
+▁release	-9.11081
+▁close	-9.11479
+rough	-9.1175
+▁similar	-9.1226
+▁French	-9.1226
+▁white	-9.12262
+▁election	-9.12265
+▁information	-9.12268
+▁employ	-9.13107
+▁influence	-9.13107
+▁capital	-9.13107
+▁regard	-9.13122
+▁children	-9.13962
+▁episode	-9.13962
+▁Greek	-9.13962
+▁return	-9.13964
+▁1990	-9.13968
+▁What	-9.13984
+▁making	-9.1399
+▁given	-9.13997
+J	-9.1477
+▁February	-9.14824
+▁amount	-9.14824
+▁psych	-9.14827
+▁application	-9.15694
+▁replac	-9.15695
+▁value	-9.15699
+▁every	-9.15723
+produc	-9.15995
+▁December	-9.16571
+▁typically	-9.16571
+▁movement	-9.16573
+▁left	-9.16578
+▁however	-9.1658
+▁previous	-9.17456
+▁claim	-9.17462
+▁Windows	-9.18349
+▁observ	-9.18349
+▁material	-9.18349
+▁object	-9.19249
+▁professional	-9.19249
+wood	-9.19263
+▁local	-9.19306
+▁trade	-9.1939
+▁Internet	-9.20159
+ought	-9.20434
+▁especial	-9.21076
+▁Western	-9.21076
+▁either	-9.21083
+pproximate	-9.22002
+▁County	-9.22002
+▁primary	-9.22936
+▁property	-9.22936
+▁account	-9.22937
+▁separate	-9.22938
+▁factor	-9.22948
+▁down	-9.22964
+▁commercial	-9.2388
+▁structure	-9.24835
+▁economy	-9.26765
+▁pressure	-9.26765
+▁attack	-9.26776
+▁condition	-9.26778
+▁decade	-9.26783
+▁week	-9.26791
+▁Confederate	-9.27745
+▁complete	-9.27745
+▁Spanish	-9.28735
+▁across	-9.28736
+▁growth	-9.28736
+▁central	-9.28738
+▁available	-9.29735
+▁certain	-9.29735
+▁2013	-9.29736
+▁You	-9.29745
+▁Whil	-9.30017
+▁primari	-9.30745
+▁contribut	-9.31766
+▁Party	-9.31768
+▁drug	-9.31777
+▁Democrat	-9.32797
+▁campaign	-9.32797
+▁subsequent	-9.32797
+▁focus	-9.32797
+▁elected	-9.32798
+craft	-9.32802
+NFL	-9.32802
+▁environment	-9.33838
+▁chemical	-9.33839
+▁entire	-9.33845
+▁radio	-9.34074
+▁William	-9.34891
+▁complex	-9.34891
+▁financial	-9.35955
+▁George	-9.35955
+▁league	-9.35955
+▁policy	-9.35956
+▁project	-9.35956
+▁compos	-9.35958
+▁litera	-9.36277
+▁physical	-9.3703
+▁historical	-9.3703
+▁hydro	-9.37031
+▁defeat	-9.37041
+▁regulat	-9.37057
+▁According	-9.38117
+▁method	-9.38117
+▁generat	-9.38195
+known	-9.39007
+▁celebrat	-9.39216
+▁built	-9.39216
+▁practice	-9.40327
+▁purpose	-9.41451
+▁credit	-9.41452
+▁affect	-9.42587
+▁degree	-9.4259
+▁beginning	-9.43736
+▁symbol	-9.43736
+▁science	-9.43738
+▁Central	-9.449
+▁industry	-9.46076
+▁private	-9.46076
+▁subject	-9.46076
+▁western	-9.46076
+▁district	-9.47266
+▁community	-9.47266
+▁problem	-9.47266
+▁determine	-9.47268
+▁Constitution	-9.48471
+▁Texas	-9.48471
+▁distinct	-9.48476
+▁except	-9.48476
+▁Association	-9.4969
+▁institution	-9.4969
+▁nuclear	-9.4969
+▁technology	-9.4969
+▁Empire	-9.49691
+▁regular	-9.49691
+▁concept	-9.49694
+▁Mexico	-9.50925
+▁announced	-9.50925
+Q	-9.69917
+q	-9.90537
+(	-9.98015