blogpost-fineweb-v1 / data /plots /wet_comparison.json
hynky's picture
hynky HF staff
update jsons
4ac9bcc
raw
history blame
9.37 kB
{"data":{"agg_score":{"ind_minhash-CC-MAIN-2019-18":{"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":[0.3308933284133672,0.35846718959510326,0.37850185949355364,0.39194786734879017,0.39600365422666073,0.40486439503729343,0.4064061753451824,0.41104014590382576,0.41393135115504265,0.41698802448809147,0.42121383734047413,0.4219294786453247,0.4234823901206255,0.42346264235675335,0.42699199728667736],"label":"Extracted from WARC"},"wet-extraction-2019-18":{"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":[0.3308620806783438,0.3493095366284251,0.367914117872715,0.37658837065100664,0.3858313206583261,0.3908915650099516,0.3968510050326586,0.39992102794349194,0.40259181894361973,0.4055726025253534,0.4074157159775495,0.40804907679557795,0.40894216299057007,0.4123705606907606,0.4108315110206604],"label":"WET data"}},"commonsense_qa/acc_norm":{"ind_minhash-CC-MAIN-2019-18":{"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":[0.2329999953508377,0.26000000536441803,0.291499987244606,0.3019999861717224,0.3034999966621399,0.3109999895095825,0.31599999964237213,0.3254999965429306,0.3210000097751617,0.3320000022649765,0.33449999988079065,0.3344999998807907,0.3334999978542328,0.3349999934434891,0.3385000079870224],"label":"Extracted from WARC"},"wet-extraction-2019-18":{"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":[0.2329999953508377,0.24249999970197675,0.2705000042915344,0.27549999952316284,0.28700000047683716,0.28449998795986176,0.29099999368190765,0.2979999929666519,0.3075000047683716,0.30550000071525574,0.3079999983310699,0.3110000044107437,0.30949999392032623,0.3114999979734421,0.3100000023841858],"label":"WET data"}},"hellaswag/acc_norm":{"ind_minhash-CC-MAIN-2019-18":{"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":[0.257999986410141,0.2915000021457672,0.33500000834465027,0.35800001025199885,0.37450000643730164,0.3859999924898147,0.3959999978542328,0.4035000056028366,0.4220000058412552,0.4294999986886978,0.43400000035762787,0.44099999964237213,0.4424999952316284,0.44449999928474426,0.4494999945163727],"label":"Extracted from WARC"},"wet-extraction-2019-18":{"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":[0.257999986410141,0.28650000691413874,0.31299999356269836,0.3305000066757202,0.3569999933242798,0.3710000067949295,0.3879999965429306,0.3854999989271164,0.39199998974800104,0.4055000096559524,0.4064999967813492,0.4065000116825104,0.4120000004768371,0.41700001060962677,0.4175000041723251],"label":"WET data"}},"openbookqa/acc_norm":{"ind_minhash-CC-MAIN-2019-18":{"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":[0.2860000133514404,0.26500000059604645,0.24800000339746475,0.28299999237060547,0.28200000524520874,0.30900000035762787,0.3100000023841858,0.3020000010728836,0.3149999976158142,0.3110000044107437,0.32100000977516174,0.31700000166893005,0.31599999964237213,0.31599999964237213,0.31900000572204584],"label":"Extracted from WARC"},"wet-extraction-2019-18":{"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":[0.2860000133514404,0.2580000013113022,0.2719999998807907,0.2770000100135803,0.27300000190734863,0.2880000025033951,0.2989999949932098,0.29500000178813934,0.29899999499320984,0.3100000023841858,0.30300000309944153,0.30600000917911524,0.3040000051259994,0.3110000044107437,0.30300000309944153],"label":"WET data"}},"piqa/acc_norm":{"ind_minhash-CC-MAIN-2019-18":{"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":[0.5099999904632568,0.6130000054836273,0.6599999964237213,0.6704999804496765,0.6845000088214874,0.6854999959468842,0.6895000040531158,0.7005000114440918,0.6990000009536743,0.7090000212192535,0.707999974489212,0.7125000059604645,0.7114999890327454,0.7094999849796295,0.7150000035762787],"label":"Extracted from WARC"},"wet-extraction-2019-18":{"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":[0.5099999904632568,0.5995000004768372,0.6385000050067902,0.6534999907016754,0.6675000190734863,0.6755000054836273,0.6814999878406525,0.6859999895095825,0.6840000152587891,0.6924999952316284,0.6944999992847443,0.69200000166893,0.6995000243186951,0.6960000097751617,0.6979999840259552],"label":"WET data"}},"winogrande/acc_norm":{"ind_minhash-CC-MAIN-2019-18":{"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":[0.4970000088214874,0.4960000067949295,0.5010000020265579,0.5055000185966492,0.5119999945163727,0.5230000019073486,0.5149999856948853,0.5180000066757202,0.5145000219345093,0.5139999985694885,0.5200000107288361,0.5140000283718109,0.5220000147819519,0.5195000171661377,0.5194999873638153],"label":"Extracted from WARC"},"wet-extraction-2019-18":{"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":[0.4970000088214874,0.49299998581409454,0.494499996304512,0.4989999979734421,0.502000018954277,0.5115000009536743,0.5119999945163727,0.5185000002384186,0.5119999945163727,0.5090000033378601,0.5175000131130219,0.5139999985694885,0.5074999928474426,0.5164999961853027,0.5115000009536743],"label":"WET data"}},"arc/acc_norm":{"ind_minhash-CC-MAIN-2019-18":{"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":[0.2509999871253967,0.2957500070333481,0.32750000059604645,0.3479999899864197,0.3422500044107437,0.3535000085830688,0.35199999809265137,0.3564999997615814,0.36150000989437103,0.36275000870227814,0.36924999952316284,0.3685000091791153,0.37325000762939453,0.3764999955892563,0.3779999911785126],"label":"Extracted from WARC"},"wet-extraction-2019-18":{"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":[0.2512499988079071,0.2689999938011169,0.304749995470047,0.32025000452995295,0.33400000631809235,0.3375000059604645,0.3384999930858612,0.346000000834465,0.34949998557567596,0.3512499928474426,0.3535000085830688,0.3577500134706497,0.35724999010562897,0.35950000584125513,0.35875000059604645],"label":"WET data"}},"mmlu/acc_norm":{"ind_minhash-CC-MAIN-2019-18":{"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":[0.2501466572284698,0.25648748874664307,0.2595148831605911,0.2695829570293426,0.27227921783924103,0.27291516959667206,0.2772494107484817,0.27682115137577057,0.2799507677555084,0.28115415573120117,0.28246068954467773,0.28493577241897583,0.2861091196537018,0.2857011407613754,0.28743599355220795],"label":"Extracted from WARC"},"wet-extraction-2019-18":{"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":[0.2501466572284698,0.2509763091802597,0.25756295025348663,0.2589569538831711,0.2636505216360092,0.2666325122117996,0.26730807125568384,0.2703682482242584,0.2727345675230026,0.273330807685852,0.2783257067203522,0.27664257586002344,0.278787299990654,0.27946445345878596,0.27840209007263184],"label":"WET data"}}},"layout":{"xaxis":{"title":{"text":"Training tokens (billions)"}},"title":{"text":"WET data is worse than data extracted from WARC"}}}