blogpost-fineweb-v1 / data /plots /custom-filters.json
hynky's picture
hynky HF staff
update slider values + fix plotly errors + update jsons to exclude sciqa
7b37a0b
raw
history blame
26.7 kB
{"data":{"agg_score":{"filtering-baseline-2019-18-40gt":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3308596294373274,0.35654734168201685,0.3758235517889261,0.38752372190356255,0.39841264486312866,0.4040419068187475,0.4097859803587198,0.41541148349642754,0.416892247274518,0.41986062191426754,0.4234193116426468,0.4218583852052688,0.4243287574499845,0.42519346065819263,0.42440339736640453],"label":"Baseline"},"filtering-custom-line-char-duplicated-v2-0.01":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3308979943394661,0.35727922804653645,0.3758955802768469,0.39312327839434147,0.3984657619148493,0.4037223849445581,0.40907647646963596,0.41408527828752995,0.42114910110831255,0.42039695382118225,0.4248786196112633,0.42590542137622833,0.4263712782412767,0.42797840014100075,0.4277621991932392],"label":"Line duplicates filter"},"filtering-custom-lines-punc-0.12":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3308933284133672,0.35735468938946724,0.3787423223257065,0.391122592613101,0.3976811040192842,0.4041402228176594,0.4110417179763317,0.4150725454092026,0.42221225984394545,0.4235249478369951,0.42567262239754194,0.42764298990368843,0.4280493911355734,0.42981273680925364,0.42845905013382435],"label":"Punctuation filter"},"filtering-custom-short-line-ratio-0.67":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.33087017294019455,0.35839469730854034,0.379800958558917,0.3909519836306572,0.3985003251582384,0.4028578344732523,0.4080309104174375,0.411550747230649,0.4152813777327537,0.41849316097795963,0.42109199613332743,0.4223319999873638,0.42558939941227436,0.42717534117400646,0.426479609683156],"label":"Short lines filter"},"filtering-custom-punc0.12-short-lines0.67-line_char_dup0.1":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3308933284133672,0.35955795273184776,0.3757704347372055,0.3934198468923569,0.398214865475893,0.4062729831784963,0.41363069601356983,0.41463132016360754,0.41851891577243805,0.4239445272833109,0.42439557053148746,0.4273625332862139,0.4289980959147215,0.4327357914298773,0.43017333932220936],"label":"Filters combined"}},"commonsense_qa/acc_norm":{"filtering-baseline-2019-18-40gt":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2329999953508377,0.2644999921321869,0.2850000113248825,0.2875000089406967,0.31049999594688416,0.3135000020265579,0.3279999941587448,0.32999999821186066,0.32349999248981476,0.3229999989271164,0.32350000739097595,0.32900001108646393,0.3264999985694885,0.3349999934434891,0.32999999821186066],"label":"Baseline"},"filtering-custom-line-char-duplicated-v2-0.01":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2329999953508377,0.2644999921321869,0.2750000059604645,0.2989999949932098,0.2974999994039535,0.31599999964237213,0.3149999976158142,0.3199999928474426,0.3244999945163727,0.3269999921321869,0.33550000190734863,0.3275000005960464,0.33599999547004694,0.3349999934434891,0.33849999308586115],"label":"Line duplicates filter"},"filtering-custom-lines-punc-0.12":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2329999953508377,0.26349999010562897,0.28849999606609344,0.29600000381469727,0.30650000274181366,0.31900000572204584,0.3229999989271164,0.3150000125169754,0.3244999945163727,0.3310000002384186,0.3310000002384186,0.32999999821186066,0.3334999978542328,0.3344999998807907,0.32999999821186066],"label":"Punctuation filter"},"filtering-custom-short-line-ratio-0.67":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2329999953508377,0.2620000094175339,0.28949999809265137,0.2974999994039535,0.30550000071525574,0.30900000035762787,0.31200000643730164,0.3190000057220459,0.32999999821186066,0.3254999965429306,0.3344999998807907,0.3320000022649765,0.3374999910593033,0.3369999974966049,0.33949999511241913],"label":"Short lines filter"},"filtering-custom-punc0.12-short-lines0.67-line_char_dup0.1":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2329999953508377,0.2649999856948852,0.2790000140666961,0.29649999737739563,0.3135000020265579,0.3164999932050705,0.32099999487400055,0.3210000097751617,0.3305000066757202,0.3205000013113022,0.32549999654293055,0.3295000046491623,0.33050000667572016,0.335999995470047,0.33200000226497645],"label":"Filters combined"}},"hellaswag/acc_norm":{"filtering-baseline-2019-18-40gt":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.257999986410141,0.2900000065565109,0.3279999941587448,0.3550000041723251,0.375,0.38850000500679016,0.40350000560283655,0.41200000047683716,0.4194999933242798,0.42249999940395355,0.4329999983310699,0.43449999392032623,0.43700000643730164,0.4395000040531158,0.43950000405311584],"label":"Baseline"},"filtering-custom-line-char-duplicated-v2-0.01":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.257999986410141,0.28900000452995295,0.3310000002384186,0.3505000025033951,0.3790000081062317,0.39250001311302185,0.40549999475479126,0.4224999994039535,0.4284999966621399,0.43050000071525574,0.43799999356269836,0.4459999948740005,0.4495000094175339,0.4564999938011169,0.4529999941587448],"label":"Line duplicates filter"},"filtering-custom-lines-punc-0.12":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.257999986410141,0.29449999332427973,0.33550000190734863,0.34800000488758087,0.3764999955892563,0.3824999928474426,0.3955000042915344,0.41799999773502344,0.4270000010728836,0.43400000035762787,0.44450001418590546,0.45049999654293055,0.45450000464916224,0.45449998974800104,0.4550000131130218],"label":"Punctuation filter"},"filtering-custom-short-line-ratio-0.67":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.257999986410141,0.3020000010728836,0.3310000002384186,0.357000008225441,0.37899999320507044,0.38850000500679016,0.3994999974966049,0.40349999070167536,0.4175000041723251,0.42400000989437103,0.4245000034570694,0.4335000067949295,0.4360000044107437,0.44750000536441803,0.44200000166893],"label":"Short lines filter"},"filtering-custom-punc0.12-short-lines0.67-line_char_dup0.1":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.257999986410141,0.29349999129772186,0.3210000097751617,0.36150000989437103,0.3734999895095825,0.39599999785423273,0.4125000089406967,0.4234999865293503,0.42749999463558197,0.44699999690055847,0.4549999982118606,0.4660000056028366,0.46600000560283655,0.47050000727176666,0.4675000011920929],"label":"Filters combined"}},"openbookqa/acc_norm":{"filtering-baseline-2019-18-40gt":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2860000133514404,0.25099999457597727,0.2629999965429306,0.2810000032186508,0.2939999997615814,0.2900000065565109,0.3100000023841858,0.3129999935626983,0.3149999976158142,0.3229999989271164,0.3310000002384186,0.32100000977516174,0.32999999821186066,0.32100000977516174,0.32100000977516174],"label":"Baseline"},"filtering-custom-line-char-duplicated-v2-0.01":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2860000133514404,0.26900000870227814,0.2670000046491623,0.306999996304512,0.2939999997615814,0.2999999970197677,0.306999996304512,0.31200000643730164,0.31299999356269836,0.3200000077486038,0.3229999989271164,0.32099999487400055,0.32500000298023224,0.3240000009536743,0.3219999969005584],"label":"Line duplicates filter"},"filtering-custom-lines-punc-0.12":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2860000133514404,0.2559999972581863,0.2849999964237213,0.3110000044107437,0.2979999929666519,0.3009999990463257,0.318000003695488,0.3140000104904175,0.32899999618530273,0.32899999618530273,0.3369999974966049,0.33599999547004694,0.32900001108646393,0.3299999982118606,0.3330000042915344],"label":"Punctuation filter"},"filtering-custom-short-line-ratio-0.67":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2860000133514404,0.25999999046325684,0.28200000524520874,0.28599999845027924,0.289000004529953,0.29999999701976776,0.31300000846385956,0.31900000572204584,0.3149999976158142,0.32099999487400055,0.3139999955892563,0.3190000057220459,0.32200001180171967,0.3229999989271164,0.3240000009536743],"label":"Short lines filter"},"filtering-custom-punc0.12-short-lines0.67-line_char_dup0.1":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2860000133514404,0.2719999998807907,0.277999997138977,0.3039999902248382,0.28199999034404755,0.30200000107288355,0.3050000071525574,0.31299999356269836,0.32099999487400055,0.3269999921321869,0.31599999964237213,0.3260000050067901,0.32600000500679016,0.3299999982118606,0.32500000298023224],"label":"Filters combined"}},"piqa/acc_norm":{"filtering-baseline-2019-18-40gt":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.5099999904632568,0.613999992609024,0.6504999995231628,0.6649999916553497,0.6870000064373016,0.6915000081062317,0.6974999904632568,0.7035000026226044,0.7129999995231628,0.7055000066757202,0.7080000042915344,0.7084999978542328,0.7114999890327454,0.714000016450882,0.7115000188350677],"label":"Baseline"},"filtering-custom-line-char-duplicated-v2-0.01":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.5099999904632568,0.609499990940094,0.652999997138977,0.6744999885559082,0.68299999833107,0.6809999942779541,0.6965000033378601,0.6995000243186951,0.7145000100135803,0.7100000083446503,0.7105000019073486,0.7134999930858612,0.7159999907016754,0.7170000076293945,0.7199999988079071],"label":"Line duplicates filter"},"filtering-custom-lines-punc-0.12":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.5099999904632568,0.6155000030994415,0.648499995470047,0.6649999916553497,0.6865000128746033,0.690500020980835,0.6965000033378601,0.7029999792575836,0.7139999866485596,0.7105000019073486,0.7089999914169312,0.7139999866485596,0.7144999802112579,0.7229999899864197,0.7175000011920929],"label":"Punctuation filter"},"filtering-custom-short-line-ratio-0.67":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.5099999904632568,0.6254999935626984,0.6530000269412994,0.6665000021457672,0.6860000193119049,0.6980000138282776,0.695499986410141,0.7084999978542328,0.7080000042915344,0.7064999938011169,0.7095000147819519,0.7129999995231628,0.7159999907016754,0.7179999947547913,0.718500018119812],"label":"Short lines filter"},"filtering-custom-punc0.12-short-lines0.67-line_char_dup0.1":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.5099999904632568,0.6274999976158142,0.656000018119812,0.6665000021457672,0.6854999959468842,0.6895000040531158,0.7035000026226044,0.7060000002384186,0.7100000083446503,0.7195000052452087,0.7159999907016754,0.715499997138977,0.7170000076293945,0.7274999916553497,0.7199999988079071],"label":"Filters combined"}},"siqa/acc_norm":{"filtering-baseline-2019-18-40gt":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.36149999499320984,0.40049999952316284,0.3959999978542328,0.39800000190734863,0.39800000190734863,0.40299999713897705,0.3969999998807907,0.40549999475479126,0.4054999947547912,0.4155000001192093,0.41800001263618464,0.40949998795986176,0.41099999845027924,0.41200000047683716,0.4065000116825104],"label":"Baseline"},"filtering-custom-line-char-duplicated-v2-0.01":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3619999885559082,0.3894999921321869,0.3989999890327453,0.4060000032186508,0.40299999713897705,0.4055000096559524,0.4095000028610229,0.40450000762939453,0.40750001370906824,0.4074999988079071,0.408500000834465,0.41050000488758087,0.40450000762939453,0.40500000119209284,0.4035000056028366],"label":"Line duplicates filter"},"filtering-custom-lines-punc-0.12":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3619999885559082,0.3930000066757202,0.39750000834465027,0.40049999952316284,0.39849999547004694,0.40449999272823334,0.4054999947547912,0.4020000100135803,0.4115000069141388,0.40800000727176666,0.402999997138977,0.4074999988079071,0.40700000524520874,0.4060000032186508,0.40250000357627863],"label":"Punctuation filter"},"filtering-custom-short-line-ratio-0.67":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3619999885559082,0.40350000560283655,0.403999999165535,0.4004999995231628,0.4010000079870224,0.39899998903274536,0.4015000015497207,0.39750000834465027,0.3969999998807907,0.4030000120401382,0.4055000096559524,0.4010000079870224,0.4020000100135803,0.40299999713897705,0.3990000039339065],"label":"Short lines filter"},"filtering-custom-punc0.12-short-lines0.67-line_char_dup0.1":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.3619999885559082,0.3889999985694885,0.4040000140666961,0.4035000056028366,0.4050000011920929,0.3995000123977661,0.4064999967813492,0.4050000011920929,0.4025000035762787,0.4055000096559524,0.40799999237060547,0.4000000059604645,0.4025000035762787,0.403999999165535,0.40150000154972076],"label":"Filters combined"}},"winogrande/acc_norm":{"filtering-baseline-2019-18-40gt":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.4970000088214874,0.4884999990463257,0.49699999392032623,0.5035000145435333,0.5054999887943268,0.5200000107288361,0.5115000009536743,0.5154999792575836,0.507999986410141,0.515500009059906,0.5160000026226044,0.5139999985694885,0.5165000259876251,0.5160000026226044,0.523499995470047],"label":"Baseline"},"filtering-custom-line-char-duplicated-v2-0.01":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.4970000088214874,0.48749999701976776,0.5024999976158142,0.5035000145435333,0.5099999904632568,0.5080000162124634,0.5050000250339508,0.5069999992847443,0.5180000066757202,0.5085000097751617,0.515500009059906,0.5165000259876251,0.5080000162124634,0.5090000033378601,0.5104999840259552],"label":"Line duplicates filter"},"filtering-custom-lines-punc-0.12":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.4970000088214874,0.48899999260902405,0.49300000071525574,0.5030000060796738,0.49799999594688416,0.5024999976158142,0.5150000154972076,0.5259999930858612,0.527999997138977,0.5245000123977661,0.5275000035762787,0.5199999809265137,0.5300000011920929,0.5300000011920929,0.5289999842643738],"label":"Punctuation filter"},"filtering-custom-short-line-ratio-0.67":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.4970000088214874,0.471000000834465,0.494499996304512,0.511000007390976,0.511000007390976,0.5070000290870667,0.50450000166893,0.5060000121593475,0.5074999928474426,0.5169999897480011,0.5264999866485596,0.526500016450882,0.5290000140666962,0.5275000035762787,0.5259999930858612],"label":"Short lines filter"},"filtering-custom-punc0.12-short-lines0.67-line_char_dup0.1":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.4970000088214874,0.4854999929666519,0.4860000014305115,0.5094999969005585,0.5090000033378601,0.5195000171661377,0.5185000002384186,0.5090000182390213,0.5084999799728394,0.5209999978542328,0.5164999961853027,0.5254999995231628,0.5250000059604645,0.5250000059604645,0.5254999995231628],"label":"Filters combined"}},"arc/acc_norm":{"filtering-baseline-2019-18-40gt":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2512499988079071,0.29224999248981476,0.3267499953508377,0.34375,0.34800000488758087,0.35224999487400055,0.3567499965429306,0.36450000107288355,0.369499996304512,0.3712500035762787,0.3722500056028366,0.37325000762939453,0.377250000834465,0.37624999880790705,0.3764999955892563],"label":"Baseline"},"filtering-custom-line-char-duplicated-v2-0.01":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2509999871253967,0.29474999010562897,0.3184999972581863,0.3392500132322311,0.35074999928474426,0.35300000011920923,0.35750000178813934,0.3684999942779541,0.3817500025033951,0.37800000607967377,0.38199999928474426,0.38600000739097595,0.38525000214576716,0.39000000059604645,0.38850000500679016],"label":"Line duplicates filter"},"filtering-custom-lines-punc-0.12":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2509999871253967,0.29100000858306885,0.31949999928474426,0.33675000071525574,0.34524999558925623,0.35850000381469727,0.3557499945163727,0.36124999821186066,0.3599999994039535,0.36800000071525574,0.36775000393390656,0.3770000040531158,0.37025000154972076,0.37424999475479126,0.37299999594688416],"label":"Punctuation filter"},"filtering-custom-short-line-ratio-0.67":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2509999871253967,0.2892500013113022,0.3190000057220459,0.3385000079870224,0.3449999988079071,0.3495000004768371,0.36374999582767487,0.3604999929666519,0.36549998819828033,0.37074999511241913,0.37150000035762787,0.3722500056028366,0.37774999439716334,0.3774999976158142,0.37899999320507044],"label":"Short lines filter"},"filtering-custom-punc0.12-short-lines0.67-line_char_dup0.1":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2509999871253967,0.2905000001192093,0.3199999928474426,0.3397499918937683,0.3467499911785126,0.3540000021457672,0.3662499934434891,0.36374999582767487,0.3647499978542328,0.3675000071525574,0.371749997138977,0.37074999511241913,0.375,0.3787499964237213,0.38099999725818634],"label":"Filters combined"}},"mmlu/acc_norm":{"filtering-baseline-2019-18-40gt":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2501270473003387,0.25162875652313227,0.26033842563629145,0.26643975079059595,0.26930116117000574,0.27358523011207575,0.27403785288333893,0.2792918980121612,0.28113801777362823,0.2826349586248398,0.2856044620275497,0.2851170748472214,0.28488004207611084,0.2877976596355438,0.28672714531421656],"label":"Baseline"},"filtering-custom-line-char-duplicated-v2-0.01":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.25018398463726044,0.2544838488101959,0.2611646503210068,0.2652362138032913,0.2704761028289795,0.2737790495157242,0.276611790060997,0.2786822021007538,0.281442791223526,0.2816756069660187,0.2860289514064789,0.28624334931373596,0.2867202013731003,0.28732720017433167,0.28609761595726013],"label":"Line duplicates filter"},"filtering-custom-lines-punc-0.12":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2501466572284698,0.2563375234603882,0.26243858039379114,0.26873072981834406,0.27219884097576136,0.27462176978588104,0.27908372879028315,0.2813303619623184,0.28369809687137604,0.28319956362247467,0.28563097119331354,0.28614395856857294,0.28564512729644775,0.2862519174814224,0.2876724004745483],"label":"Punctuation filter"},"filtering-custom-short-line-ratio-0.67":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.24996141344308848,0.25390757620334625,0.26540763676166534,0.27061584591865534,0.27150256931781763,0.2718626409769058,0.27449728548526764,0.2784059643745422,0.28175103664398193,0.28019529581069946,0.2827359586954117,0.2814059555530548,0.2844651788473129,0.28390273451805115,0.2838368713855743],"label":"Short lines filter"},"filtering-custom-punc0.12-short-lines0.67-line_char_dup0.1":{"x":[0.0,2.0971520000000003,4.194304000000001,6.291456,8.388608000000001,10.48576,12.582912,14.680064000000002,16.777216000000003,18.874368,20.97152,23.068672000000003,25.165824,27.262976000000002,28.311552000000002],"y":[0.2501466572284698,0.2534636557102203,0.2621634304523468,0.2661087810993194,0.2704689502716064,0.27318383753299713,0.2757955640554428,0.2758005559444427,0.28340134024620056,0.2835562080144882,0.28641459345817566,0.28565025329589844,0.28998473286628723,0.29013633728027344,0.2888867110013962],"label":"Filters combined"}}},"layout":{"xaxis":{"title":{"text":"Training tokens (Billions)"}},"title":{"text":"Custom filters performance"},"yaxis":{"range":[0.38,0.44]}},"defaultWindowSize":3}