{"data":{"agg_score":{"cross_minhash_dump_CC-MAIN-2013-48":{"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":[0.3310184087604284,0.3494944926351309,0.3678930029273033,0.3791136778891086,0.3830251954495907,0.387223158031702,0.3940111547708511,0.3980898857116699,0.398512527346611,0.3974943198263645,0.4026404283940792,0.402598962187767,0.4074418470263481,0.4055770002305507,0.4050002694129944],"label":"Originally kept data"},"deduped_removed_cross":{"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":[0.3308933284133672,0.3570646308362484,0.3725825920701027,0.383445743471384,0.39065178856253624,0.3996846079826355,0.4021379072219133,0.4061895925551653,0.41160152666270733,0.4141362868249416,0.4196407739073038,0.4217643104493618,0.4209167677909136,0.42394610680639744,0.4236117731779814],"label":"Originally removed data"}},"commonsense_qa/acc_norm":{"cross_minhash_dump_CC-MAIN-2013-48":{"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":[0.2329999953508377,0.2479999959468841,0.2800000011920929,0.2910000085830688,0.289000004529953,0.3059999942779541,0.3050000071525574,0.3050000071525574,0.3149999976158142,0.3140000104904175,0.3269999921321869,0.3219999969005584,0.3190000057220459,0.3179999887943268,0.3120000064373016],"label":"Originally kept data"},"deduped_removed_cross":{"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":[0.2329999953508377,0.2580000013113022,0.28649999201297754,0.289000004529953,0.29500000178813934,0.30949999392032623,0.31599999964237213,0.31700000166893005,0.318000003695488,0.32549999654293055,0.32099999487400055,0.33250001072883606,0.32500000298023224,0.3330000042915344,0.32750000059604645],"label":"Originally removed data"}},"hellaswag/acc_norm":{"cross_minhash_dump_CC-MAIN-2013-48":{"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":[0.257999986410141,0.2860000133514404,0.2879999876022339,0.328000009059906,0.3379999995231628,0.356000006198883,0.356000006198883,0.3589999973773956,0.3720000088214874,0.3740000128746032,0.382999986410141,0.3810000121593475,0.395000010728836,0.3849999904632568,0.3930000066757202],"label":"Originally kept data"},"deduped_removed_cross":{"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":[0.257999986410141,0.2775000035762787,0.3085000067949295,0.32750000059604645,0.35600000619888306,0.36999998986721033,0.37950000166893005,0.3965000063180923,0.41050000488758087,0.41250000894069666,0.42149999737739563,0.4270000010728836,0.4339999854564667,0.4389999955892563,0.4375],"label":"Originally removed data"}},"openbookqa/acc_norm":{"cross_minhash_dump_CC-MAIN-2013-48":{"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":[0.2860000133514404,0.2440000027418136,0.2800000011920929,0.2660000026226043,0.2800000011920929,0.2759999930858612,0.2879999876022339,0.3019999861717224,0.2879999876022339,0.2739999890327453,0.2800000011920929,0.2840000092983246,0.2899999916553497,0.2899999916553497,0.2879999876022339],"label":"Originally kept data"},"deduped_removed_cross":{"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":[0.2860000133514404,0.26899999380111694,0.26899999380111694,0.2870000004768371,0.2849999964237213,0.29200001060962677,0.29900000989437103,0.29900000989437103,0.2980000078678131,0.2939999997615814,0.32199999690055847,0.31700000166893005,0.30799999833106995,0.31700000166893005,0.3260000050067901],"label":"Originally removed data"}},"piqa/acc_norm":{"cross_minhash_dump_CC-MAIN-2013-48":{"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":[0.5099999904632568,0.6140000224113464,0.6480000019073486,0.6539999842643738,0.6669999957084656,0.6549999713897705,0.6679999828338623,0.6850000023841858,0.671999990940094,0.6869999766349792,0.6840000152587891,0.6880000233650208,0.6890000104904175,0.6980000138282776,0.6940000057220459],"label":"Originally kept data"},"deduped_removed_cross":{"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":[0.5099999904632568,0.6155000030994415,0.6419999897480011,0.6620000004768372,0.6780000030994415,0.6819999814033508,0.6865000128746033,0.6884999871253967,0.7064999938011169,0.7080000042915344,0.7055000066757202,0.7114999890327454,0.715499997138977,0.7084999978542328,0.7074999809265137],"label":"Originally removed data"}},"winogrande/acc_norm":{"cross_minhash_dump_CC-MAIN-2013-48":{"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":[0.4970000088214874,0.492000013589859,0.4990000128746032,0.5040000081062317,0.4959999918937683,0.5109999775886536,0.5210000276565552,0.5099999904632568,0.5080000162124634,0.5059999823570251,0.5130000114440918,0.515999972820282,0.5099999904632568,0.5099999904632568,0.5130000114440918],"label":"Originally kept data"},"deduped_removed_cross":{"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":[0.4970000088214874,0.49900001287460327,0.50450000166893,0.4985000044107437,0.502499982714653,0.5170000195503235,0.5030000060796738,0.5210000276565552,0.5160000026226044,0.5200000107288361,0.527999997138977,0.5224999785423279,0.5259999930858612,0.5240000188350677,0.5214999914169312],"label":"Originally removed data"}},"arc/acc_norm":{"cross_minhash_dump_CC-MAIN-2013-48":{"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":[0.2520000040531158,0.277999997138977,0.3115000128746032,0.3334999978542328,0.3375000059604645,0.3379999995231628,0.351500004529953,0.3549999892711639,0.3630000054836273,0.3610000014305115,0.3650000095367431,0.3659999966621399,0.3700000047683716,0.3729999959468841,0.3659999966621399],"label":"Originally kept data"},"deduped_removed_cross":{"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":[0.2509999871253967,0.29625000059604645,0.3167499899864197,0.3412500023841858,0.3410000056028366,0.35625000298023224,0.3604999929666519,0.36050000786781305,0.3652499914169311,0.3722499907016754,0.37925000488758087,0.3789999932050705,0.3782500028610229,0.38750000298023224,0.3879999965429306],"label":"Originally removed data"}},"mmlu/acc_norm":{"cross_minhash_dump_CC-MAIN-2013-48":{"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":[0.25014728307724,0.2539559006690979,0.2536440193653106,0.2624094188213348,0.2637015581130981,0.2667853236198425,0.2655892074108124,0.2677191197872162,0.2741002142429352,0.271954596042633,0.2721233963966369,0.2767916917800903,0.2795347571372986,0.2746160328388214,0.2780021429061889],"label":"Originally kept data"},"deduped_removed_cross":{"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":[0.2501466572284698,0.2537670284509659,0.25891076028347015,0.26481594145298004,0.27071431279182434,0.2712268680334091,0.27510324120521545,0.2755167037248611,0.28106220066547394,0.28384028375148773,0.2848761975765228,0.2871145009994507,0.28658416867256165,0.288568839430809,0.2888942211866379],"label":"Originally removed data"}}},"layout":{"xaxis":{"title":{"text":"Training tokens (billions)"}},"title":{"text":"The originally removed data outperforms the kept data"}}}