daa-tokenizers / tokenizer_results.json
Ouaill's picture
Upload tokenizer_results.json with huggingface_hub
8943545 verified
Raw
History Blame Contribute Delete
20.8 kB
[
{
"name":"shared_bpe_8000",
"tokenizer_type":"shared",
"algorithm":"BPE",
"vocab_size":8000,
"fertility_overall":1.7787163674,
"cpt_overall":3.0049111942,
"fertility_disparity":0.2198677878,
"cpt_disparity":0.645823621,
"oov_disparity":0.0,
"vocab_gini":0.7709496335,
"shannon_entropy":10.5561176463,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1755236174,
"morph_consistency_precision":0.1706382107,
"morph_consistency_recall":0.1556559587,
"morph_consistency_f1":0.161694338,
"ar_fertility":1.656426646,
"ar_cpt":3.3838216268,
"ar_oov_rate":0.0,
"ar_mean_seq_len":17.6595461798,
"ar_median_seq_len":17.0,
"az_fertility":1.8762944338,
"az_cpt":2.7379980058,
"az_oov_rate":0.0,
"az_mean_seq_len":25.9109200496,
"az_median_seq_len":26.0
},
{
"name":"concat_bpe_8000",
"tokenizer_type":"concatenated",
"algorithm":"BPE",
"vocab_size":8000,
"fertility_overall":1.6889873506,
"cpt_overall":3.1645498835,
"fertility_disparity":0.1648715075,
"cpt_disparity":0.0316881659,
"oov_disparity":0.0,
"vocab_gini":0.6849702921,
"shannon_entropy":10.9244878509,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.2157843708,
"morph_consistency_precision":0.1569809449,
"morph_consistency_recall":0.1856069998,
"morph_consistency_f1":0.1679847615,
"ar_fertility":1.7806883265,
"ar_cpt":3.1476885788,
"ar_oov_rate":0.0,
"ar_mean_seq_len":19.1354369793,
"ar_median_seq_len":19.0,
"az_fertility":1.615816819,
"az_cpt":3.1793767446,
"az_oov_rate":0.0,
"az_mean_seq_len":22.036429711,
"az_median_seq_len":22.0
},
{
"name":"shared_unigram_8000",
"tokenizer_type":"shared",
"algorithm":"Unigram",
"vocab_size":8000,
"fertility_overall":1.7790411363,
"cpt_overall":3.0043626393,
"fertility_disparity":0.1474708507,
"cpt_disparity":0.5176722057,
"oov_disparity":0.0,
"vocab_gini":0.8189437499,
"shannon_entropy":10.165482266,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1823029664,
"morph_consistency_precision":0.1498923892,
"morph_consistency_recall":0.2196576099,
"morph_consistency_f1":0.1770370715,
"ar_fertility":1.6970183469,
"ar_cpt":3.3028825634,
"ar_oov_rate":0.0,
"ar_mean_seq_len":18.1438574721,
"ar_median_seq_len":18.0,
"az_fertility":1.8444891977,
"az_cpt":2.7852103577,
"az_oov_rate":0.0,
"az_mean_seq_len":25.4391065414,
"az_median_seq_len":25.0
},
{
"name":"concat_unigram_8000",
"tokenizer_type":"concatenated",
"algorithm":"Unigram",
"vocab_size":8000,
"fertility_overall":1.6947867958,
"cpt_overall":3.1537210089,
"fertility_disparity":0.1669302238,
"cpt_disparity":0.0343315623,
"oov_disparity":0.0,
"vocab_gini":0.7623152124,
"shannon_entropy":10.4233732784,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.2111496914,
"morph_consistency_precision":0.1450557842,
"morph_consistency_recall":0.2404900444,
"morph_consistency_f1":0.1800784056,
"ar_fertility":1.7876328228,
"ar_cpt":3.1354606137,
"ar_oov_rate":0.0,
"ar_mean_seq_len":19.2194646339,
"ar_median_seq_len":19.0,
"az_fertility":1.620702599,
"az_cpt":3.169792176,
"az_oov_rate":0.0,
"az_mean_seq_len":22.1099982273,
"az_median_seq_len":22.0
},
{
"name":"shared_wordpiece_8000",
"tokenizer_type":"shared",
"algorithm":"WordPiece",
"vocab_size":8000,
"fertility_overall":1.7761712394,
"cpt_overall":3.0092170198,
"fertility_disparity":0.226757188,
"cpt_disparity":0.6596486988,
"oov_disparity":0.0,
"vocab_gini":0.7564900789,
"shannon_entropy":10.5708706321,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1765983213,
"morph_consistency_precision":0.163672929,
"morph_consistency_recall":0.1493781435,
"morph_consistency_f1":0.1551907022,
"ar_fertility":1.6500496569,
"ar_cpt":3.3968991687,
"ar_oov_rate":0.0,
"ar_mean_seq_len":17.5850912959,
"ar_median_seq_len":17.0,
"az_fertility":1.8768068449,
"az_cpt":2.7372504699,
"az_oov_rate":0.0,
"az_mean_seq_len":25.9190746322,
"az_median_seq_len":25.0
},
{
"name":"concat_wordpiece_8000",
"tokenizer_type":"concatenated",
"algorithm":"WordPiece",
"vocab_size":8000,
"fertility_overall":1.6962184303,
"cpt_overall":3.1510592198,
"fertility_disparity":0.1560458869,
"cpt_disparity":0.0140028965,
"oov_disparity":0.0,
"vocab_gini":0.6600277207,
"shannon_entropy":10.9155665053,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.219108083,
"morph_consistency_precision":0.1522420473,
"morph_consistency_recall":0.175816502,
"morph_consistency_f1":0.1618608459,
"ar_fertility":1.7830106258,
"ar_cpt":3.1435888416,
"ar_oov_rate":0.0,
"ar_mean_seq_len":19.163623471,
"ar_median_seq_len":19.0,
"az_fertility":1.626964739,
"az_cpt":3.1575917381,
"az_oov_rate":0.0,
"az_mean_seq_len":22.2024463748,
"az_median_seq_len":22.0
},
{
"name":"shared_bbpe_8000",
"tokenizer_type":"shared",
"algorithm":"BBPE",
"vocab_size":8000,
"fertility_overall":1.9554271209,
"cpt_overall":2.7333592066,
"fertility_disparity":0.4718091571,
"cpt_disparity":0.9376231111,
"oov_disparity":0.0,
"vocab_gini":0.8121558079,
"shannon_entropy":9.9528287353,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1921228413,
"morph_consistency_precision":0.1550132007,
"morph_consistency_recall":0.2827125976,
"morph_consistency_f1":0.1995081577,
"ar_fertility":1.6930084603,
"ar_cpt":3.310705433,
"ar_oov_rate":0.0,
"ar_mean_seq_len":18.0962595285,
"ar_median_seq_len":18.0,
"az_fertility":2.1648176174,
"az_cpt":2.3730823219,
"az_oov_rate":0.0,
"az_mean_seq_len":30.2043963836,
"az_median_seq_len":30.0
},
{
"name":"concat_bbpe_8000",
"tokenizer_type":"concatenated",
"algorithm":"BBPE",
"vocab_size":8000,
"fertility_overall":1.8827749848,
"cpt_overall":2.8388335126,
"fertility_disparity":0.116792759,
"cpt_disparity":0.4279320309,
"oov_disparity":0.0,
"vocab_gini":0.7491107045,
"shannon_entropy":10.2263454675,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.2316542111,
"morph_consistency_precision":0.1588581248,
"morph_consistency_recall":0.3282934754,
"morph_consistency_f1":0.2135590301,
"ar_fertility":1.8178152465,
"ar_cpt":3.0834004272,
"ar_oov_rate":0.0,
"ar_mean_seq_len":19.5777344442,
"ar_median_seq_len":19.0,
"az_fertility":1.9346080055,
"az_cpt":2.6554683963,
"az_oov_rate":0.0,
"az_mean_seq_len":26.7797376352,
"az_median_seq_len":26.0
},
{
"name":"shared_bpe_16000",
"tokenizer_type":"shared",
"algorithm":"BPE",
"vocab_size":16000,
"fertility_overall":1.5971771615,
"cpt_overall":3.3464570197,
"fertility_disparity":0.1510951681,
"cpt_disparity":0.6173754613,
"oov_disparity":0.0,
"vocab_gini":0.7920106017,
"shannon_entropy":11.090783206,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1164562127,
"morph_consistency_precision":0.1645828627,
"morph_consistency_recall":0.111253923,
"morph_consistency_f1":0.1323611335,
"ar_fertility":1.513138539,
"ar_cpt":3.7042558651,
"ar_oov_rate":0.0,
"ar_mean_seq_len":15.9586952668,
"ar_median_seq_len":16.0,
"az_fertility":1.6642337071,
"az_cpt":3.0868804038,
"az_oov_rate":0.0,
"az_mean_seq_len":22.756248892,
"az_median_seq_len":22.0
},
{
"name":"concat_bpe_16000",
"tokenizer_type":"concatenated",
"algorithm":"BPE",
"vocab_size":16000,
"fertility_overall":1.5165781285,
"cpt_overall":3.5243055556,
"fertility_disparity":0.1426774709,
"cpt_disparity":0.0229367001,
"oov_disparity":0.0,
"vocab_gini":0.7224875042,
"shannon_entropy":11.453646644,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1531217,
"morph_consistency_precision":0.1543768971,
"morph_consistency_recall":0.1249030296,
"morph_consistency_f1":0.1376491323,
"ar_fertility":1.5959348561,
"ar_cpt":3.5120808886,
"ar_oov_rate":0.0,
"ar_mean_seq_len":16.9423861018,
"ar_median_seq_len":17.0,
"az_fertility":1.4532573853,
"az_cpt":3.5350175887,
"az_oov_rate":0.0,
"az_mean_seq_len":19.6181528098,
"az_median_seq_len":19.0
},
{
"name":"shared_unigram_16000",
"tokenizer_type":"shared",
"algorithm":"Unigram",
"vocab_size":16000,
"fertility_overall":1.6212862838,
"cpt_overall":3.296693975,
"fertility_disparity":0.0885280481,
"cpt_disparity":0.4717666206,
"oov_disparity":0.0,
"vocab_gini":0.8334856469,
"shannon_entropy":10.5776002094,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1315210046,
"morph_consistency_precision":0.1557524276,
"morph_consistency_recall":0.1706022629,
"morph_consistency_f1":0.1622742892,
"ar_fertility":1.5720472823,
"ar_cpt":3.5654476623,
"ar_oov_rate":0.0,
"ar_mean_seq_len":16.6604325474,
"ar_median_seq_len":16.0,
"az_fertility":1.6605753304,
"az_cpt":3.0936810417,
"az_oov_rate":0.0,
"az_mean_seq_len":22.7031554689,
"az_median_seq_len":22.0
},
{
"name":"concat_unigram_16000",
"tokenizer_type":"concatenated",
"algorithm":"Unigram",
"vocab_size":16000,
"fertility_overall":1.541731151,
"cpt_overall":3.4668072447,
"fertility_disparity":0.1464437495,
"cpt_disparity":0.0256827057,
"oov_disparity":0.0,
"vocab_gini":0.7918225207,
"shannon_entropy":10.8508668291,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1582156543,
"morph_consistency_precision":0.1464142753,
"morph_consistency_recall":0.1741476553,
"morph_consistency_f1":0.1574918159,
"ar_fertility":1.6231826701,
"ar_cpt":3.4531247844,
"ar_oov_rate":0.0,
"ar_mean_seq_len":17.2674171246,
"ar_median_seq_len":17.0,
"az_fertility":1.4767389206,
"az_cpt":3.4788074901,
"az_oov_rate":0.0,
"az_mean_seq_len":19.9683566743,
"az_median_seq_len":20.0
},
{
"name":"shared_wordpiece_16000",
"tokenizer_type":"shared",
"algorithm":"WordPiece",
"vocab_size":16000,
"fertility_overall":1.5914307397,
"cpt_overall":3.3585405825,
"fertility_disparity":0.1511682691,
"cpt_disparity":0.6209619713,
"oov_disparity":0.0,
"vocab_gini":0.7797479401,
"shannon_entropy":11.1061302427,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1158134243,
"morph_consistency_precision":0.1653640001,
"morph_consistency_recall":0.1209851544,
"morph_consistency_f1":0.1385333865,
"ar_fertility":1.5073514587,
"ar_cpt":3.7184773832,
"ar_oov_rate":0.0,
"ar_mean_seq_len":15.8912426875,
"ar_median_seq_len":16.0,
"az_fertility":1.6585197278,
"az_cpt":3.0975154119,
"az_oov_rate":0.0,
"az_mean_seq_len":22.6717780535,
"az_median_seq_len":22.0
},
{
"name":"concat_wordpiece_16000",
"tokenizer_type":"concatenated",
"algorithm":"WordPiece",
"vocab_size":16000,
"fertility_overall":1.5161705103,
"cpt_overall":3.5252530551,
"fertility_disparity":0.1360666065,
"cpt_disparity":0.0077904651,
"oov_disparity":0.0,
"vocab_gini":0.7024364505,
"shannon_entropy":11.4443519139,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1506488459,
"morph_consistency_precision":0.1740477945,
"morph_consistency_recall":0.1384490501,
"morph_consistency_f1":0.1533202072,
"ar_fertility":1.5918502976,
"ar_cpt":3.5210926029,
"ar_oov_rate":0.0,
"ar_mean_seq_len":16.8945222478,
"ar_median_seq_len":17.0,
"az_fertility":1.455783691,
"az_cpt":3.528883068,
"az_oov_rate":0.0,
"az_mean_seq_len":19.6559120723,
"az_median_seq_len":19.0
},
{
"name":"shared_bbpe_16000",
"tokenizer_type":"shared",
"algorithm":"BBPE",
"vocab_size":16000,
"fertility_overall":1.7936656802,
"cpt_overall":2.9798667514,
"fertility_disparity":0.4264991353,
"cpt_disparity":1.0104466245,
"oov_disparity":0.0,
"vocab_gini":0.8352512331,
"shannon_entropy":10.2895041219,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1365421097,
"morph_consistency_precision":0.154622604,
"morph_consistency_recall":0.2260888365,
"morph_consistency_f1":0.1823875169,
"ar_fertility":1.5564483008,
"ar_cpt":3.6011811666,
"ar_oov_rate":0.0,
"ar_mean_seq_len":16.4752703421,
"ar_median_seq_len":16.0,
"az_fertility":1.9829474362,
"az_cpt":2.590734542,
"az_oov_rate":0.0,
"az_mean_seq_len":27.498847722,
"az_median_seq_len":27.0
},
{
"name":"concat_bbpe_16000",
"tokenizer_type":"concatenated",
"algorithm":"BBPE",
"vocab_size":16000,
"fertility_overall":1.7340440692,
"cpt_overall":3.0823234649,
"fertility_disparity":0.1722144964,
"cpt_disparity":0.5838072031,
"oov_disparity":0.0,
"vocab_gini":0.7908187337,
"shannon_entropy":10.5217570863,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1719709073,
"morph_consistency_precision":0.1620182352,
"morph_consistency_recall":0.2599866609,
"morph_consistency_f1":0.1986209129,
"ar_fertility":1.6382589476,
"ar_cpt":3.4213469831,
"ar_oov_rate":0.0,
"ar_mean_seq_len":17.4463747562,
"ar_median_seq_len":17.0,
"az_fertility":1.810473444,
"az_cpt":2.83753978,
"az_oov_rate":0.0,
"az_mean_seq_len":24.9330792413,
"az_median_seq_len":24.0
},
{
"name":"shared_bpe_32000",
"tokenizer_type":"shared",
"algorithm":"BPE",
"vocab_size":32000,
"fertility_overall":1.4562440141,
"cpt_overall":3.670322193,
"fertility_disparity":0.0726954505,
"cpt_disparity":0.507593217,
"oov_disparity":0.0,
"vocab_gini":0.7969114377,
"shannon_entropy":11.521256996,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.0705763007,
"morph_consistency_precision":0.1610095066,
"morph_consistency_recall":0.0899012584,
"morph_consistency_f1":0.1149172148,
"ar_fertility":1.41581105,
"ar_cpt":3.9588985464,
"ar_oov_rate":0.0,
"ar_mean_seq_len":14.8034036518,
"ar_median_seq_len":14.0,
"az_fertility":1.4885065005,
"az_cpt":3.4513053294,
"az_oov_rate":0.0,
"az_mean_seq_len":20.1420847367,
"az_median_seq_len":20.0
},
{
"name":"concat_bpe_32000",
"tokenizer_type":"concatenated",
"algorithm":"BPE",
"vocab_size":32000,
"fertility_overall":1.3911576687,
"cpt_overall":3.8420409447,
"fertility_disparity":0.134007069,
"cpt_disparity":0.0335692487,
"oov_disparity":0.0,
"vocab_gini":0.7567616479,
"shannon_entropy":11.8182442917,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.0970004442,
"morph_consistency_precision":0.148728366,
"morph_consistency_recall":0.0942436549,
"morph_consistency_f1":0.1146174635,
"ar_fertility":1.4656919481,
"ar_cpt":3.8241680422,
"ar_oov_rate":0.0,
"ar_mean_seq_len":15.3963836199,
"ar_median_seq_len":15.0,
"az_fertility":1.3316848791,
"az_cpt":3.8577372909,
"az_oov_rate":0.0,
"az_mean_seq_len":17.8096082255,
"az_median_seq_len":17.0
},
{
"name":"shared_unigram_32000",
"tokenizer_type":"shared",
"algorithm":"Unigram",
"vocab_size":32000,
"fertility_overall":1.5033984749,
"cpt_overall":3.55520164,
"fertility_disparity":0.019300727,
"cpt_disparity":0.3573083439,
"oov_disparity":0.0,
"vocab_gini":0.8324888295,
"shannon_entropy":10.9257151273,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.0979264521,
"morph_consistency_precision":0.1631020109,
"morph_consistency_recall":0.1383463815,
"morph_consistency_f1":0.1482409685,
"ar_fertility":1.4926634757,
"ar_cpt":3.7550676352,
"ar_oov_rate":0.0,
"ar_mean_seq_len":15.7181350824,
"ar_median_seq_len":15.0,
"az_fertility":1.5119642027,
"az_cpt":3.3977592913,
"az_oov_rate":0.0,
"az_mean_seq_len":20.4923772381,
"az_median_seq_len":20.0
},
{
"name":"concat_unigram_32000",
"tokenizer_type":"concatenated",
"algorithm":"Unigram",
"vocab_size":32000,
"fertility_overall":1.4304050001,
"cpt_overall":3.7366233501,
"fertility_disparity":0.1449358921,
"cpt_disparity":0.0511471793,
"oov_disparity":0.0,
"vocab_gini":0.8069023024,
"shannon_entropy":11.1617375452,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1111640248,
"morph_consistency_precision":0.1486025565,
"morph_consistency_recall":0.1393090166,
"morph_consistency_f1":0.1420302247,
"ar_fertility":1.5110178541,
"ar_cpt":3.709454718,
"ar_oov_rate":0.0,
"ar_mean_seq_len":15.9360042546,
"ar_median_seq_len":16.0,
"az_fertility":1.3660819619,
"az_cpt":3.7606018973,
"az_oov_rate":0.0,
"az_mean_seq_len":18.3221946463,
"az_median_seq_len":18.0
},
{
"name":"shared_wordpiece_32000",
"tokenizer_type":"shared",
"algorithm":"WordPiece",
"vocab_size":32000,
"fertility_overall":1.4501065441,
"cpt_overall":3.6858565637,
"fertility_disparity":0.0723742616,
"cpt_disparity":0.5097029774,
"oov_disparity":0.0,
"vocab_gini":0.7885550388,
"shannon_entropy":11.5290013722,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.0695593122,
"morph_consistency_precision":0.1813894046,
"morph_consistency_recall":0.0999073414,
"morph_consistency_f1":0.1273030756,
"ar_fertility":1.4098522241,
"ar_cpt":3.975631071,
"ar_oov_rate":0.0,
"ar_mean_seq_len":14.7339124269,
"ar_median_seq_len":14.0,
"az_fertility":1.4822264857,
"az_cpt":3.4659280936,
"az_oov_rate":0.0,
"az_mean_seq_len":20.0491934054,
"az_median_seq_len":20.0
},
{
"name":"concat_wordpiece_32000",
"tokenizer_type":"concatenated",
"algorithm":"WordPiece",
"vocab_size":32000,
"fertility_overall":1.3885462613,
"cpt_overall":3.8492665836,
"fertility_disparity":0.1308885617,
"cpt_disparity":0.0257554455,
"oov_disparity":0.0,
"vocab_gini":0.7418783914,
"shannon_entropy":11.7951971085,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.097541522,
"morph_consistency_precision":0.1635636627,
"morph_consistency_recall":0.112578089,
"morph_consistency_f1":0.1328802305,
"ar_fertility":1.4613460375,
"ar_cpt":3.8355407712,
"ar_oov_rate":0.0,
"ar_mean_seq_len":15.3454174792,
"ar_median_seq_len":15.0,
"az_fertility":1.3304574758,
"az_cpt":3.8612962167,
"az_oov_rate":0.0,
"az_mean_seq_len":17.7915263251,
"az_median_seq_len":17.0
},
{
"name":"shared_bbpe_32000",
"tokenizer_type":"shared",
"algorithm":"BBPE",
"vocab_size":32000,
"fertility_overall":1.673696699,
"cpt_overall":3.1934607548,
"fertility_disparity":0.3736950757,
"cpt_disparity":1.0310609307,
"oov_disparity":0.0,
"vocab_gini":0.843090999,
"shannon_entropy":10.5235434613,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.0973786272,
"morph_consistency_precision":0.157092523,
"morph_consistency_recall":0.1976111484,
"morph_consistency_f1":0.1737146088,
"ar_fertility":1.4658487593,
"ar_cpt":3.8237589465,
"ar_oov_rate":0.0,
"ar_mean_seq_len":15.3998404538,
"ar_median_seq_len":15.0,
"az_fertility":1.839543835,
"az_cpt":2.7926980158,
"az_oov_rate":0.0,
"az_mean_seq_len":25.3655380252,
"az_median_seq_len":25.0
},
{
"name":"concat_bbpe_32000",
"tokenizer_type":"concatenated",
"algorithm":"BBPE",
"vocab_size":32000,
"fertility_overall":1.6299390561,
"cpt_overall":3.2791929879,
"fertility_disparity":0.2066556668,
"cpt_disparity":0.7157815641,
"oov_disparity":0.0,
"vocab_gini":0.8217315637,
"shannon_entropy":10.6841474757,
"exact_match_rate":0.99,
"morph_edit_distance_ar":4.1215903942,
"morph_consistency_precision":0.1574322292,
"morph_consistency_recall":0.2143887572,
"morph_consistency_f1":0.1797621337,
"ar_fertility":1.5149978718,
"ar_cpt":3.6997096906,
"ar_oov_rate":0.0,
"ar_mean_seq_len":15.9832476511,
"ar_median_seq_len":16.0,
"az_fertility":1.7216535386,
"az_cpt":2.9839281265,
"az_oov_rate":0.0,
"az_mean_seq_len":23.6117709626,
"az_median_seq_len":23.0
}
]