上传交互图表数据

.gitattributes

@@ -41,3 +41,4 @@ _astro/placeholder.BA7Wxfbp_StF8u.webp filter=lfs diff=lfs merge=lfs -text
 _astro/placeholder.BA7Wxfbp.png filter=lfs diff=lfs merge=lfs -text
 _astro/smoll-training-guide.V1MW5_f8.png filter=lfs diff=lfs merge=lfs -text
 _astro/visual-vocabulary-poster.-2uGKfDN.png filter=lfs diff=lfs merge=lfs -text
+data/data.json filter=lfs diff=lfs merge=lfs -text

data/admin_seq_write.json

@@ -0,0 +1,402 @@
+{
+  "metadata": {
+    "thread_counts": [
+      1,
+      4,
+      8,
+      16,
+      32,
+      64
+    ],
+    "io_size_labels": [
+      "64K",
+      "128K",
+      "256K",
+      "512K",
+      "1M",
+      "2M",
+      "4M",
+      "8M"
+    ]
+  },
+  "data": {
+    "CPUONLY": {
+      "throughput": [
+        [
+          0.152469,
+          0.26117,
+          0.414248,
+          0.578068,
+          0.565348,
+          0.568225,
+          0.577506,
+          0.572973
+        ],
+        [
+          0.540033,
+          0.789152,
+          1.086652,
+          1.104094,
+          1.088801,
+          1.080391,
+          1.083259,
+          1.08143
+        ],
+        [
+          0.836049,
+          1.076984,
+          1.053262,
+          1.123994,
+          1.129315,
+          1.117614,
+          1.118362,
+          1.1194
+        ],
+        [
+          0.994074,
+          1.060973,
+          1.120867,
+          1.12639,
+          1.12329,
+          1.120485,
+          1.107331,
+          1.087737
+        ],
+        [
+          0.993441,
+          1.087584,
+          1.090756,
+          1.106212,
+          1.122596,
+          1.095533,
+          1.101319,
+          1.069289
+        ],
+        [
+          1.025168,
+          1.087341,
+          1.092217,
+          1.12571,
+          1.121396,
+          1.126329,
+          1.068233,
+          1.021527
+        ]
+      ],
+      "iops": [
+        [
+          2498.0479070234733,
+          2139.507210892529,
+          1696.7602905701997,
+          1183.882839321853,
+          578.9163692378943,
+          290.93110566432773,
+          147.84143627955348,
+          73.34050557245995
+        ],
+        [
+          8847.897159784274,
+          6464.730336991869,
+          4450.928140341348,
+          2261.184153791194,
+          1114.9320195002053,
+          553.1601178181616,
+          277.3143560830297,
+          138.42306675252544
+        ],
+        [
+          13697.819939847417,
+          8822.654733941103,
+          4314.160849506365,
+          2301.9400319271776,
+          1156.418317666152,
+          572.2183901666812,
+          286.3006671177363,
+          143.2832282960037
+        ],
+        [
+          16286.914382392424,
+          8691.492323112161,
+          4591.069704152401,
+          2306.8476622185954,
+          1150.2488842925661,
+          573.6884152922094,
+          283.47668141029067,
+          139.23039516175982
+        ],
+        [
+          16276.534349062445,
+          8909.491224246749,
+          4467.737016627545,
+          2265.5228944736855,
+          1149.5384638418893,
+          560.91306472439,
+          281.93762421142344,
+          136.869006273676
+        ],
+        [
+          16796.36045857987,
+          8907.497796313595,
+          4473.722318493241,
+          2305.4531163602633,
+          1148.309366337562,
+          576.6803947732119,
+          273.4676275065704,
+          130.7554220851424
+        ]
+      ]
+    },
+    "CPU_GPU": {
+      "throughput": [
+        [
+          0.133594,
+          0.219962,
+          0.306266,
+          0.39437,
+          0.392275,
+          0.382734,
+          0.391264,
+          0.382689
+        ],
+        [
+          0.447735,
+          0.658977,
+          0.827244,
+          0.889111,
+          0.847647,
+          0.770023,
+          0.706995,
+          0.620418
+        ],
+        [
+          0.724719,
+          1.014614,
+          0.983895,
+          1.011178,
+          1.005857,
+          0.916667,
+          0.840629,
+          0.696714
+        ],
+        [
+          0.983819,
+          1.063914,
+          1.100767,
+          1.068469,
+          0.999162,
+          0.923699,
+          0.839825,
+          0.683286
+        ],
+        [
+          1.005289,
+          1.103092,
+          1.116685,
+          1.053272,
+          1.005367,
+          0.896358,
+          0.852786,
+          0.684293
+        ],
+        [
+          1.065023,
+          1.08159,
+          1.11725,
+          1.055289,
+          0.944936,
+          1.116448,
+          1.056321,
+          1.026407
+        ]
+      ],
+      "iops": [
+        [
+          2188.805003810641,
+          1801.9327244518913,
+          1254.4648112275472,
+          807.6702789633604,
+          401.6892156942314,
+          195.95988131478288,
+          100.1635788316634,
+          48.98414542896463
+        ],
+        [
+          7335.69547493148,
+          5398.342821071309,
+          3388.3902548920755,
+          1820.8983859176906,
+          867.9908622691727,
+          394.2517232266254,
+          180.990597871184,
+          79.41355210191702
+        ],
+        [
+          11873.799843095047,
+          8311.719442546639,
+          4030.033346349065,
+          2070.893281226892,
+          1029.9974699700138,
+          469.3336846882585,
+          215.20098305035026,
+          89.17936160568651
+        ],
+        [
+          16118.897728079322,
+          8715.582079315056,
+          4508.740071863426,
+          2188.224544121712,
+          1023.1416233518968,
+          472.93399638020134,
+          214.9951285031559,
+          87.46058838832784
+        ],
+        [
+          16470.66019205259,
+          9036.527450496105,
+          4573.942259026739,
+          2157.1012487306243,
+          1029.4962439564893,
+          458.93536722733324,
+          218.3132410462806,
+          87.58947749978103
+        ],
+        [
+          17449.329554326698,
+          8860.382418682879,
+          4576.256901227057,
+          2161.2325929268327,
+          967.6141443988694,
+          571.6216206650973,
+          270.418082324846,
+          131.38005086254503
+        ]
+      ]
+    },
+    "GPUD": {
+      "throughput": [
+        [
+          0.142515,
+          0.231796,
+          0.382949,
+          0.504668,
+          0.481376,
+          0.485502,
+          0.48984,
+          0.515456
+        ],
+        [
+          0.496469,
+          0.77063,
+          1.08494,
+          1.112721,
+          1.107903,
+          1.081599,
+          1.072657,
+          1.077407
+        ],
+        [
+          0.840859,
+          1.035756,
+          1.064612,
+          1.116211,
+          1.115864,
+          1.109883,
+          1.094589,
+          1.098854
+        ],
+        [
+          1.003986,
+          1.07968,
+          1.124783,
+          1.123465,
+          1.08724,
+          1.092136,
+          1.088111,
+          1.098928
+        ],
+        [
+          1.014784,
+          1.10672,
+          1.129549,
+          1.130887,
+          1.12454,
+          1.121108,
+          1.083096,
+          1.054424
+        ],
+        [
+          1.058534,
+          1.068297,
+          1.115003,
+          1.118295,
+          1.103522,
+          1.071732,
+          1.062585,
+          1.027981
+        ]
+      ],
+      "iops": [
+        [
+          2334.9678289098874,
+          1898.8711253545282,
+          1568.5598406424333,
+          1033.5597798568836,
+          492.9293324185811,
+          248.5767698601948,
+          125.39911097255285,
+          65.97832813462294
+        ],
+        [
+          8134.147854630187,
+          6312.999208989576,
+          4443.91412662719,
+          2278.8535004982828,
+          1134.493161750125,
+          553.7788602849582,
+          274.6000659802936,
+          137.90813157270262
+        ],
+        [
+          13776.63618195407,
+          8484.910697289442,
+          4360.6492564990385,
+          2286.0009153096144,
+          1142.6452045193573,
+          568.2598751884161,
+          280.21466219207684,
+          140.65331254179182
+        ],
+        [
+          16449.31203716111,
+          8844.738653898388,
+          4607.109946917802,
+          2300.8561884386445,
+          1113.3336377759263,
+          559.1735157357342,
+          278.5564570916211,
+          140.6627208136504
+        ],
+        [
+          16626.215956993066,
+          9066.2538412566,
+          4626.633973595312,
+          2316.055719667878,
+          1151.5292376106368,
+          574.0074110954939,
+          277.27266135782764,
+          134.96630039478194
+        ],
+        [
+          17343.028788774318,
+          8751.48565529327,
+          4567.052839727334,
+          2290.267706916213,
+          1130.006746293577,
+          548.7268252653923,
+          272.02184521733426,
+          131.58159694457942
+        ]
+      ]
+    }
+  }
+}

data/attention_evals.csv

@@ -0,0 +1,1338 @@
+run_name,metric,tokens,value
+GQA 8 groups (baseline),hellaswag,629145600.0,0.2493527185819558
+GQA 8 groups (baseline),hellaswag,786432000.0,0.25184226249751046
+GQA 8 groups (baseline),hellaswag,943718400.0,0.25632344154550885
+GQA 8 groups (baseline),hellaswag,1101004800.0,0.2620991834295957
+GQA 8 groups (baseline),hellaswag,1258291200.0,0.26319458275243973
+GQA 8 groups (baseline),hellaswag,1572864000.0,0.26618203545110536
+GQA 8 groups (baseline),hellaswag,3145728000.0,0.30193188607847044
+GQA 8 groups (baseline),hellaswag,4718592000.0,0.32971519617606054
+GQA 8 groups (baseline),hellaswag,6291456000.0,0.35062736506671976
+GQA 8 groups (baseline),hellaswag,7864320000.0,0.36735710017924716
+GQA 8 groups (baseline),hellaswag,9437184000.0,0.3842859988050189
+GQA 8 groups (baseline),hellaswag,11010048000.0,0.3948416650069707
+GQA 8 groups (baseline),hellaswag,12582912000.0,0.4056960764787891
+GQA 8 groups (baseline),hellaswag,14155776000.0,0.41575383389762993
+GQA 8 groups (baseline),hellaswag,15728640000.0,0.4250149372634933
+GQA 8 groups (baseline),hellaswag,17301504000.0,0.43308105954989046
+GQA 8 groups (baseline),hellaswag,18874368000.0,0.4393547102170882
+GQA 8 groups (baseline),hellaswag,20447232000.0,0.4418442541326429
+GQA 8 groups (baseline),hellaswag,22020096000.0,0.4462258514240191
+GQA 8 groups (baseline),hellaswag,23592960000.0,0.4554869547898825
+GQA 8 groups (baseline),hellaswag,25165824000.0,0.46016729735112527
+GQA 8 groups (baseline),hellaswag,26738688000.0,0.465345548695479
+GQA 8 groups (baseline),hellaswag,28311552000.0,0.47012547301334395
+GQA 8 groups (baseline),hellaswag,29884416000.0,0.4766978689504083
+GQA 8 groups (baseline),hellaswag,31457280000.0,0.47918741286596295
+GQA 8 groups (baseline),hellaswag,33030144000.0,0.48078072097191793
+GQA 8 groups (baseline),hellaswag,34603008000.0,0.4865564628560048
+GQA 8 groups (baseline),hellaswag,36175872000.0,0.4876518621788488
+GQA 8 groups (baseline),hellaswag,37748736000.0,0.49044015136427005
+GQA 8 groups (baseline),hellaswag,39321600000.0,0.49063931487751444
+GQA 8 groups (baseline),hellaswag,40894464000.0,0.4931288587930691
+GQA 8 groups (baseline),hellaswag,42467328000.0,0.4918342959569807
+GQA 8 groups (baseline),hellaswag,44040192000.0,0.49512049392551283
+GQA 8 groups (baseline),hellaswag,45613056000.0,0.5008962358095996
+GQA 8 groups (baseline),hellaswag,47185920000.0,0.5011949810794663
+GQA 8 groups (baseline),hellaswag,47185920000.0,0.5011949810794663
+GQA 8 groups (baseline),mmlu,629145600.0,0.2489163499384853
+GQA 8 groups (baseline),mmlu,786432000.0,0.2517603616264398
+GQA 8 groups (baseline),mmlu,943718400.0,0.25724213780689
+GQA 8 groups (baseline),mmlu,1101004800.0,0.25680173747753837
+GQA 8 groups (baseline),mmlu,1258291200.0,0.25669502347341705
+GQA 8 groups (baseline),mmlu,1572864000.0,0.2529772632728836
+GQA 8 groups (baseline),mmlu,3145728000.0,0.266739436414485
+GQA 8 groups (baseline),mmlu,4718592000.0,0.2761609150042153
+GQA 8 groups (baseline),mmlu,6291456000.0,0.27730228072464913
+GQA 8 groups (baseline),mmlu,7864320000.0,0.2887388551122367
+GQA 8 groups (baseline),mmlu,9437184000.0,0.2953111113032989
+GQA 8 groups (baseline),mmlu,11010048000.0,0.2970070415346663
+GQA 8 groups (baseline),mmlu,12582912000.0,0.3007672634359463
+GQA 8 groups (baseline),mmlu,14155776000.0,0.30561499053592006
+GQA 8 groups (baseline),mmlu,15728640000.0,0.3041190871315019
+GQA 8 groups (baseline),mmlu,17301504000.0,0.30460641017653367
+GQA 8 groups (baseline),mmlu,18874368000.0,0.31062281378537093
+GQA 8 groups (baseline),mmlu,20447232000.0,0.31986412164024264
+GQA 8 groups (baseline),mmlu,22020096000.0,0.3123236868278878
+GQA 8 groups (baseline),mmlu,23592960000.0,0.316372229668946
+GQA 8 groups (baseline),mmlu,25165824000.0,0.3139885571253008
+GQA 8 groups (baseline),mmlu,26738688000.0,0.3162923575086662
+GQA 8 groups (baseline),mmlu,28311552000.0,0.31967181375719017
+GQA 8 groups (baseline),mmlu,29884416000.0,0.32451565902328705
+GQA 8 groups (baseline),mmlu,31457280000.0,0.32343607991224915
+GQA 8 groups (baseline),mmlu,33030144000.0,0.32311396751649274
+GQA 8 groups (baseline),mmlu,34603008000.0,0.328688062201764
+GQA 8 groups (baseline),mmlu,36175872000.0,0.32595165000892123
+GQA 8 groups (baseline),mmlu,37748736000.0,0.32130001597034497
+GQA 8 groups (baseline),mmlu,39321600000.0,0.3287989794742651
+GQA 8 groups (baseline),mmlu,40894464000.0,0.3253808507449651
+GQA 8 groups (baseline),mmlu,42467328000.0,0.32714305078178496
+GQA 8 groups (baseline),mmlu,44040192000.0,0.32491375376743536
+GQA 8 groups (baseline),mmlu,45613056000.0,0.3284839336158802
+GQA 8 groups (baseline),mmlu,47185920000.0,0.33255698746721873
+GQA 8 groups (baseline),mmlu,47185920000.0,0.33255698746721873
+GQA 8 groups (baseline),arc,629145600.0,0.2565443973293803
+GQA 8 groups (baseline),arc,786432000.0,0.26943280932188785
+GQA 8 groups (baseline),arc,943718400.0,0.2660255857781455
+GQA 8 groups (baseline),arc,1101004800.0,0.2715027119890601
+GQA 8 groups (baseline),arc,1258291200.0,0.28014789533560863
+GQA 8 groups (baseline),arc,1572864000.0,0.2887471127658841
+GQA 8 groups (baseline),arc,3145728000.0,0.3456917008538169
+GQA 8 groups (baseline),arc,4718592000.0,0.3634546546235966
+GQA 8 groups (baseline),arc,6291456000.0,0.3896638455085554
+GQA 8 groups (baseline),arc,7864320000.0,0.39786517047609204
+GQA 8 groups (baseline),arc,9437184000.0,0.40465663460544005
+GQA 8 groups (baseline),arc,11010048000.0,0.4021428735592558
+GQA 8 groups (baseline),arc,12582912000.0,0.4285337734569816
+GQA 8 groups (baseline),arc,14155776000.0,0.4338579193527999
+GQA 8 groups (baseline),arc,15728640000.0,0.4484355787683433
+GQA 8 groups (baseline),arc,17301504000.0,0.43825412831385524
+GQA 8 groups (baseline),arc,18874368000.0,0.4573371657416026
+GQA 8 groups (baseline),arc,20447232000.0,0.46049373139816824
+GQA 8 groups (baseline),arc,22020096000.0,0.452667315935234
+GQA 8 groups (baseline),arc,23592960000.0,0.45706352489628943
+GQA 8 groups (baseline),arc,25165824000.0,0.445835631629147
+GQA 8 groups (baseline),arc,26738688000.0,0.4612665333655095
+GQA 8 groups (baseline),arc,28311552000.0,0.4629672722676136
+GQA 8 groups (baseline),arc,29884416000.0,0.4652878328219625
+GQA 8 groups (baseline),arc,31457280000.0,0.4716526757909011
+GQA 8 groups (baseline),arc,33030144000.0,0.4716526757909011
+GQA 8 groups (baseline),arc,34603008000.0,0.48025189322117645
+GQA 8 groups (baseline),arc,36175872000.0,0.4707994334700819
+GQA 8 groups (baseline),arc,37748736000.0,0.4714479838199975
+GQA 8 groups (baseline),arc,39321600000.0,0.47228973466174834
+GQA 8 groups (baseline),arc,40894464000.0,0.48265289412900336
+GQA 8 groups (baseline),arc,42467328000.0,0.48094066374783095
+GQA 8 groups (baseline),arc,44040192000.0,0.4840972294043967
+GQA 8 groups (baseline),arc,45613056000.0,0.48431341285436846
+GQA 8 groups (baseline),arc,47185920000.0,0.48373381137886257
+GQA 8 groups (baseline),arc,47185920000.0,0.48373381137886257
+GQA 8 groups (baseline),piqa,629145600.0,0.5272034820457019
+GQA 8 groups (baseline),piqa,786432000.0,0.5250272034820457
+GQA 8 groups (baseline),piqa,943718400.0,0.5359085963003264
+GQA 8 groups (baseline),piqa,1101004800.0,0.5424374319912949
+GQA 8 groups (baseline),piqa,1258291200.0,0.558215451577802
+GQA 8 groups (baseline),piqa,1572864000.0,0.5712731229597389
+GQA 8 groups (baseline),piqa,3145728000.0,0.5973884657236126
+GQA 8 groups (baseline),piqa,4718592000.0,0.6153427638737758
+GQA 8 groups (baseline),piqa,6291456000.0,0.6349292709466812
+GQA 8 groups (baseline),piqa,7864320000.0,0.6452665941240479
+GQA 8 groups (baseline),piqa,9437184000.0,0.6583242655059848
+GQA 8 groups (baseline),piqa,11010048000.0,0.6610446137105549
+GQA 8 groups (baseline),piqa,12582912000.0,0.6708378672470077
+GQA 8 groups (baseline),piqa,14155776000.0,0.6659412404787813
+GQA 8 groups (baseline),piqa,15728640000.0,0.6822633297062024
+GQA 8 groups (baseline),piqa,17301504000.0,0.6789989118607181
+GQA 8 groups (baseline),piqa,18874368000.0,0.6789989118607181
+GQA 8 groups (baseline),piqa,20447232000.0,0.7007616974972797
+GQA 8 groups (baseline),piqa,22020096000.0,0.6893362350380848
+GQA 8 groups (baseline),piqa,23592960000.0,0.6936887921653971
+GQA 8 groups (baseline),piqa,25165824000.0,0.6898803046789989
+GQA 8 groups (baseline),piqa,26738688000.0,0.690424374319913
+GQA 8 groups (baseline),piqa,28311552000.0,0.6947769314472253
+GQA 8 groups (baseline),piqa,29884416000.0,0.6947769314472253
+GQA 8 groups (baseline),piqa,31457280000.0,0.7002176278563657
+GQA 8 groups (baseline),piqa,33030144000.0,0.6964091403699674
+GQA 8 groups (baseline),piqa,34603008000.0,0.705658324265506
+GQA 8 groups (baseline),piqa,36175872000.0,0.6958650707290533
+GQA 8 groups (baseline),piqa,37748736000.0,0.6980413492927094
+GQA 8 groups (baseline),piqa,39321600000.0,0.7029379760609358
+GQA 8 groups (baseline),piqa,40894464000.0,0.7089227421109902
+GQA 8 groups (baseline),piqa,42467328000.0,0.7029379760609358
+GQA 8 groups (baseline),piqa,44040192000.0,0.705114254624592
+GQA 8 groups (baseline),piqa,45613056000.0,0.7110990206746464
+GQA 8 groups (baseline),piqa,47185920000.0,0.7105549510337323
+GQA 8 groups (baseline),piqa,47185920000.0,0.7105549510337323
+GQA 8 groups (baseline),commonsenseqa,629145600.0,0.2244062244062244
+GQA 8 groups (baseline),commonsenseqa,786432000.0,0.23177723177723178
+GQA 8 groups (baseline),commonsenseqa,943718400.0,0.23177723177723178
+GQA 8 groups (baseline),commonsenseqa,1101004800.0,0.21621621621621623
+GQA 8 groups (baseline),commonsenseqa,1258291200.0,0.2416052416052416
+GQA 8 groups (baseline),commonsenseqa,1572864000.0,0.25634725634725636
+GQA 8 groups (baseline),commonsenseqa,3145728000.0,0.2964782964782965
+GQA 8 groups (baseline),commonsenseqa,4718592000.0,0.2964782964782965
+GQA 8 groups (baseline),commonsenseqa,6291456000.0,0.32432432432432434
+GQA 8 groups (baseline),commonsenseqa,7864320000.0,0.343980343980344
+GQA 8 groups (baseline),commonsenseqa,9437184000.0,0.3579033579033579
+GQA 8 groups (baseline),commonsenseqa,11010048000.0,0.33497133497133497
+GQA 8 groups (baseline),commonsenseqa,12582912000.0,0.3677313677313677
+GQA 8 groups (baseline),commonsenseqa,14155776000.0,0.3783783783783784
+GQA 8 groups (baseline),commonsenseqa,15728640000.0,0.36855036855036855
+GQA 8 groups (baseline),commonsenseqa,17301504000.0,0.3939393939393939
+GQA 8 groups (baseline),commonsenseqa,18874368000.0,0.4013104013104013
+GQA 8 groups (baseline),commonsenseqa,20447232000.0,0.3906633906633907
+GQA 8 groups (baseline),commonsenseqa,22020096000.0,0.38656838656838655
+GQA 8 groups (baseline),commonsenseqa,23592960000.0,0.3939393939393939
+GQA 8 groups (baseline),commonsenseqa,25165824000.0,0.3570843570843571
+GQA 8 groups (baseline),commonsenseqa,26738688000.0,0.38493038493038495
+GQA 8 groups (baseline),commonsenseqa,28311552000.0,0.3832923832923833
+GQA 8 groups (baseline),commonsenseqa,29884416000.0,0.3857493857493858
+GQA 8 groups (baseline),commonsenseqa,31457280000.0,0.3931203931203931
+GQA 8 groups (baseline),commonsenseqa,33030144000.0,0.39803439803439805
+GQA 8 groups (baseline),commonsenseqa,34603008000.0,0.3988533988533989
+GQA 8 groups (baseline),commonsenseqa,36175872000.0,0.38984438984438985
+GQA 8 groups (baseline),commonsenseqa,37748736000.0,0.39967239967239965
+GQA 8 groups (baseline),commonsenseqa,39321600000.0,0.38738738738738737
+GQA 8 groups (baseline),commonsenseqa,40894464000.0,0.3988533988533989
+GQA 8 groups (baseline),commonsenseqa,42467328000.0,0.39803439803439805
+GQA 8 groups (baseline),commonsenseqa,44040192000.0,0.40294840294840295
+GQA 8 groups (baseline),commonsenseqa,45613056000.0,0.4037674037674038
+GQA 8 groups (baseline),commonsenseqa,47185920000.0,0.39967239967239965
+GQA 8 groups (baseline),commonsenseqa,47185920000.0,0.39967239967239965
+GQA 8 groups (baseline),winogrande,629145600.0,0.47355958958168903
+GQA 8 groups (baseline),winogrande,786432000.0,0.49013417521704816
+GQA 8 groups (baseline),winogrande,943718400.0,0.4861878453038674
+GQA 8 groups (baseline),winogrande,1101004800.0,0.4909234411996843
+GQA 8 groups (baseline),winogrande,1258291200.0,0.48303078137332284
+GQA 8 groups (baseline),winogrande,1572864000.0,0.48697711128650356
+GQA 8 groups (baseline),winogrande,3145728000.0,0.4925019731649566
+GQA 8 groups (baseline),winogrande,4718592000.0,0.5074980268350434
+GQA 8 groups (baseline),winogrande,6291456000.0,0.49329123914759276
+GQA 8 groups (baseline),winogrande,7864320000.0,0.4909234411996843
+GQA 8 groups (baseline),winogrande,9437184000.0,0.48539857932123126
+GQA 8 groups (baseline),winogrande,11010048000.0,0.489344909234412
+GQA 8 groups (baseline),winogrande,12582912000.0,0.489344909234412
+GQA 8 groups (baseline),winogrande,14155776000.0,0.4972375690607735
+GQA 8 groups (baseline),winogrande,15728640000.0,0.48855564325177586
+GQA 8 groups (baseline),winogrande,17301504000.0,0.4996053670086819
+GQA 8 groups (baseline),winogrande,18874368000.0,0.4925019731649566
+GQA 8 groups (baseline),winogrande,20447232000.0,0.500394632991318
+GQA 8 groups (baseline),winogrande,22020096000.0,0.4964483030781373
+GQA 8 groups (baseline),winogrande,23592960000.0,0.4956590370955012
+GQA 8 groups (baseline),winogrande,25165824000.0,0.500394632991318
+GQA 8 groups (baseline),winogrande,26738688000.0,0.4940805051302289
+GQA 8 groups (baseline),winogrande,28311552000.0,0.4980268350434096
+GQA 8 groups (baseline),winogrande,29884416000.0,0.4988161010260458
+GQA 8 groups (baseline),winogrande,31457280000.0,0.4925019731649566
+GQA 8 groups (baseline),winogrande,33030144000.0,0.5011838989739542
+GQA 8 groups (baseline),winogrande,34603008000.0,0.5035516969218626
+GQA 8 groups (baseline),winogrande,36175872000.0,0.4925019731649566
+GQA 8 groups (baseline),winogrande,37748736000.0,0.5067087608524072
+GQA 8 groups (baseline),winogrande,39321600000.0,0.5019731649565904
+GQA 8 groups (baseline),winogrande,40894464000.0,0.505130228887135
+GQA 8 groups (baseline),winogrande,42467328000.0,0.5043409629044988
+GQA 8 groups (baseline),winogrande,44040192000.0,0.500394632991318
+GQA 8 groups (baseline),winogrande,45613056000.0,0.5082872928176796
+GQA 8 groups (baseline),winogrande,47185920000.0,0.5059194948697711
+GQA 8 groups (baseline),winogrande,47185920000.0,0.5059194948697711
+GQA 8 groups (baseline),openbookqa,629145600.0,0.25
+GQA 8 groups (baseline),openbookqa,786432000.0,0.268
+GQA 8 groups (baseline),openbookqa,943718400.0,0.274
+GQA 8 groups (baseline),openbookqa,1101004800.0,0.286
+GQA 8 groups (baseline),openbookqa,1258291200.0,0.252
+GQA 8 groups (baseline),openbookqa,1572864000.0,0.248
+GQA 8 groups (baseline),openbookqa,3145728000.0,0.288
+GQA 8 groups (baseline),openbookqa,4718592000.0,0.26
+GQA 8 groups (baseline),openbookqa,6291456000.0,0.282
+GQA 8 groups (baseline),openbookqa,7864320000.0,0.29
+GQA 8 groups (baseline),openbookqa,9437184000.0,0.28
+GQA 8 groups (baseline),openbookqa,11010048000.0,0.302
+GQA 8 groups (baseline),openbookqa,12582912000.0,0.29
+GQA 8 groups (baseline),openbookqa,14155776000.0,0.298
+GQA 8 groups (baseline),openbookqa,15728640000.0,0.31
+GQA 8 groups (baseline),openbookqa,17301504000.0,0.308
+GQA 8 groups (baseline),openbookqa,18874368000.0,0.318
+GQA 8 groups (baseline),openbookqa,20447232000.0,0.318
+GQA 8 groups (baseline),openbookqa,22020096000.0,0.326
+GQA 8 groups (baseline),openbookqa,23592960000.0,0.312
+GQA 8 groups (baseline),openbookqa,25165824000.0,0.314
+GQA 8 groups (baseline),openbookqa,26738688000.0,0.3
+GQA 8 groups (baseline),openbookqa,28311552000.0,0.324
+GQA 8 groups (baseline),openbookqa,29884416000.0,0.296
+GQA 8 groups (baseline),openbookqa,31457280000.0,0.314
+GQA 8 groups (baseline),openbookqa,33030144000.0,0.31
+GQA 8 groups (baseline),openbookqa,34603008000.0,0.314
+GQA 8 groups (baseline),openbookqa,36175872000.0,0.316
+GQA 8 groups (baseline),openbookqa,37748736000.0,0.322
+GQA 8 groups (baseline),openbookqa,39321600000.0,0.318
+GQA 8 groups (baseline),openbookqa,40894464000.0,0.318
+GQA 8 groups (baseline),openbookqa,42467328000.0,0.316
+GQA 8 groups (baseline),openbookqa,44040192000.0,0.316
+GQA 8 groups (baseline),openbookqa,45613056000.0,0.32
+GQA 8 groups (baseline),openbookqa,47185920000.0,0.314
+GQA 8 groups (baseline),openbookqa,47185920000.0,0.314
+MHA,hellaswag,1572864000.0,0.2634933280223063
+MHA,hellaswag,3145728000.0,0.29954192391953793
+MHA,hellaswag,4718592000.0,0.32443736307508464
+MHA,hellaswag,6291456000.0,0.3450507866958773
+MHA,hellaswag,7864320000.0,0.3681537542322247
+MHA,hellaswag,9437184000.0,0.38099980083648677
+MHA,hellaswag,11010048000.0,0.39673371838279226
+MHA,hellaswag,12582912000.0,0.40201155148376816
+MHA,hellaswag,14155776000.0,0.4166500697072296
+MHA,hellaswag,15728640000.0,0.42272455686118304
+MHA,hellaswag,17301504000.0,0.4331806413065126
+MHA,hellaswag,18874368000.0,0.43766182035451107
+MHA,hellaswag,20447232000.0,0.44293965345548697
+MHA,hellaswag,22020096000.0,0.4473212507468632
+MHA,hellaswag,23592960000.0,0.4554869547898825
+MHA,hellaswag,25165824000.0,0.46046604262099183
+MHA,hellaswag,26738688000.0,0.4675363473411671
+MHA,hellaswag,28311552000.0,0.46773551085441145
+MHA,hellaswag,29884416000.0,0.4714200358494324
+MHA,hellaswag,31457280000.0,0.47769368651663013
+MHA,hellaswag,33030144000.0,0.48008364867556264
+MHA,hellaswag,34603008000.0,0.4825731925911173
+MHA,hellaswag,36175872000.0,0.48615813582951606
+MHA,hellaswag,37748736000.0,0.4879506074487154
+MHA,hellaswag,39321600000.0,0.49432383987253536
+MHA,hellaswag,40894464000.0,0.4945230033857797
+MHA,hellaswag,42467328000.0,0.4945230033857797
+MHA,hellaswag,44040192000.0,0.498406691894045
+MHA,hellaswag,45613056000.0,0.49790878311093406
+MHA,hellaswag,47185920000.0,0.4992033459470225
+MHA,hellaswag,47185920000.0,0.4992033459470225
+MHA,mmlu,1572864000.0,0.2538503284440106
+MHA,mmlu,3145728000.0,0.2660650899710644
+MHA,mmlu,4718592000.0,0.2775894262176124
+MHA,mmlu,6291456000.0,0.28056070742116956
+MHA,mmlu,7864320000.0,0.28706082429952784
+MHA,mmlu,9437184000.0,0.2905202778120329
+MHA,mmlu,11010048000.0,0.2970895059107717
+MHA,mmlu,12582912000.0,0.29885544083886817
+MHA,mmlu,14155776000.0,0.30410952737856956
+MHA,mmlu,15728640000.0,0.3066997123471385
+MHA,mmlu,17301504000.0,0.3110110782120057
+MHA,mmlu,18874368000.0,0.30370244045627914
+MHA,mmlu,20447232000.0,0.3146673244623488
+MHA,mmlu,22020096000.0,0.30838034810714887
+MHA,mmlu,23592960000.0,0.3108493853447417
+MHA,mmlu,25165824000.0,0.31187059184075583
+MHA,mmlu,26738688000.0,0.3146522772616844
+MHA,mmlu,28311552000.0,0.31721283505138786
+MHA,mmlu,29884416000.0,0.31417940704367203
+MHA,mmlu,31457280000.0,0.32282606732608
+MHA,mmlu,33030144000.0,0.3240958666818058
+MHA,mmlu,34603008000.0,0.32417330866954897
+MHA,mmlu,36175872000.0,0.32229839730524906
+MHA,mmlu,37748736000.0,0.32150151331312726
+MHA,mmlu,39321600000.0,0.3253164605257358
+MHA,mmlu,40894464000.0,0.3281675331084834
+MHA,mmlu,42467328000.0,0.329159476577291
+MHA,mmlu,44040192000.0,0.32703307410727533
+MHA,mmlu,45613056000.0,0.3290723842808839
+MHA,mmlu,47185920000.0,0.32680301675775253
+MHA,mmlu,47185920000.0,0.32680301675775253
+MHA,arc,1572864000.0,0.27688790636742855
+MHA,arc,3145728000.0,0.3408343962951471
+MHA,arc,4718592000.0,0.36284632445042
+MHA,arc,6291456000.0,0.3835611519058618
+MHA,arc,7864320000.0,0.38746753657163213
+MHA,arc,9437184000.0,0.40733486744578895
+MHA,arc,11010048000.0,0.4096899024373427
+MHA,arc,12582912000.0,0.41853403201526074
+MHA,arc,14155776000.0,0.4337487503016513
+MHA,arc,15728640000.0,0.4289259201801864
+MHA,arc,17301504000.0,0.42463097987842013
+MHA,arc,18874368000.0,0.42954574183243127
+MHA,arc,20447232000.0,0.4362682570873697
+MHA,arc,22020096000.0,0.43591633054090395
+MHA,arc,23592960000.0,0.440079118833385
+MHA,arc,25165824000.0,0.45282604198986454
+MHA,arc,26738688000.0,0.4465911963778858
+MHA,arc,28311552000.0,0.45749589179623307
+MHA,arc,29884416000.0,0.448115972006757
+MHA,arc,31457280000.0,0.45785930982176715
+MHA,arc,33030144000.0,0.46083416646556574
+MHA,arc,34603008000.0,0.4705602670619735
+MHA,arc,36175872000.0,0.46362156835706325
+MHA,arc,37748736000.0,0.4697644821364958
+MHA,arc,39321600000.0,0.4689169855552108
+MHA,arc,40894464000.0,0.4690757116098413
+MHA,arc,42467328000.0,0.4695253157283874
+MHA,arc,44040192000.0,0.47714703922041807
+MHA,arc,45613056000.0,0.47738620562852646
+MHA,arc,47185920000.0,0.4744113489847278
+MHA,arc,47185920000.0,0.4744113489847278
+MHA,piqa,1572864000.0,0.5614798694232862
+MHA,piqa,3145728000.0,0.5957562568008705
+MHA,piqa,4718592000.0,0.6229597388465724
+MHA,piqa,6291456000.0,0.6365614798694232
+MHA,piqa,7864320000.0,0.6452665941240479
+MHA,piqa,9437184000.0,0.6610446137105549
+MHA,piqa,11010048000.0,0.6713819368879217
+MHA,piqa,12582912000.0,0.6735582154515778
+MHA,piqa,14155776000.0,0.6708378672470077
+MHA,piqa,15728640000.0,0.6795429815016322
+MHA,piqa,17301504000.0,0.6849836779107725
+MHA,piqa,18874368000.0,0.6789989118607181
+MHA,piqa,20447232000.0,0.6953210010881393
+MHA,piqa,22020096000.0,0.6877040261153428
+MHA,piqa,23592960000.0,0.6893362350380848
+MHA,piqa,25165824000.0,0.690968443960827
+MHA,piqa,26738688000.0,0.6980413492927094
+MHA,piqa,28311552000.0,0.6974972796517954
+MHA,piqa,29884416000.0,0.6996735582154516
+MHA,piqa,31457280000.0,0.6947769314472253
+MHA,piqa,33030144000.0,0.7083786724700761
+MHA,piqa,34603008000.0,0.705114254624592
+MHA,piqa,36175872000.0,0.7083786724700761
+MHA,piqa,37748736000.0,0.7089227421109902
+MHA,piqa,39321600000.0,0.7083786724700761
+MHA,piqa,40894464000.0,0.7121871599564744
+MHA,piqa,42467328000.0,0.7105549510337323
+MHA,piqa,44040192000.0,0.7072905331882481
+MHA,piqa,45613056000.0,0.7132752992383025
+MHA,piqa,47185920000.0,0.7127312295973884
+MHA,piqa,47185920000.0,0.7127312295973884
+MHA,commonsenseqa,1572864000.0,0.24488124488124488
+MHA,commonsenseqa,3145728000.0,0.29074529074529076
+MHA,commonsenseqa,4718592000.0,0.3194103194103194
+MHA,commonsenseqa,6291456000.0,0.33906633906633904
+MHA,commonsenseqa,7864320000.0,0.3357903357903358
+MHA,commonsenseqa,9437184000.0,0.34807534807534807
+MHA,commonsenseqa,11010048000.0,0.34807534807534807
+MHA,commonsenseqa,12582912000.0,0.3669123669123669
+MHA,commonsenseqa,14155776000.0,0.3783783783783784
+MHA,commonsenseqa,15728640000.0,0.3775593775593776
+MHA,commonsenseqa,17301504000.0,0.3628173628173628
+MHA,commonsenseqa,18874368000.0,0.36855036855036855
+MHA,commonsenseqa,20447232000.0,0.36936936936936937
+MHA,commonsenseqa,22020096000.0,0.3783783783783784
+MHA,commonsenseqa,23592960000.0,0.411957411957412
+MHA,commonsenseqa,25165824000.0,0.3882063882063882
+MHA,commonsenseqa,26738688000.0,0.3923013923013923
+MHA,commonsenseqa,28311552000.0,0.37919737919737917
+MHA,commonsenseqa,29884416000.0,0.3841113841113841
+MHA,commonsenseqa,31457280000.0,0.3923013923013923
+MHA,commonsenseqa,33030144000.0,0.38984438984438985
+MHA,commonsenseqa,34603008000.0,0.40786240786240785
+MHA,commonsenseqa,36175872000.0,0.3972153972153972
+MHA,commonsenseqa,37748736000.0,0.4152334152334152
+MHA,commonsenseqa,39321600000.0,0.3963963963963964
+MHA,commonsenseqa,40894464000.0,0.4013104013104013
+MHA,commonsenseqa,42467328000.0,0.40294840294840295
+MHA,commonsenseqa,44040192000.0,0.4045864045864046
+MHA,commonsenseqa,45613056000.0,0.41277641277641275
+MHA,commonsenseqa,47185920000.0,0.39803439803439805
+MHA,commonsenseqa,47185920000.0,0.39803439803439805
+MHA,winogrande,1572864000.0,0.48539857932123126
+MHA,winogrande,3145728000.0,0.47829518547750594
+MHA,winogrande,4718592000.0,0.4877663772691397
+MHA,winogrande,6291456000.0,0.5035516969218626
+MHA,winogrande,7864320000.0,0.4909234411996843
+MHA,winogrande,9437184000.0,0.4956590370955012
+MHA,winogrande,11010048000.0,0.4925019731649566
+MHA,winogrande,12582912000.0,0.505130228887135
+MHA,winogrande,14155776000.0,0.4980268350434096
+MHA,winogrande,15728640000.0,0.4940805051302289
+MHA,winogrande,17301504000.0,0.48224151539068666
+MHA,winogrande,18874368000.0,0.4846093133385951
+MHA,winogrande,20447232000.0,0.5019731649565904
+MHA,winogrande,22020096000.0,0.48539857932123126
+MHA,winogrande,23592960000.0,0.48697711128650356
+MHA,winogrande,25165824000.0,0.4940805051302289
+MHA,winogrande,26738688000.0,0.48539857932123126
+MHA,winogrande,28311552000.0,0.5043409629044988
+MHA,winogrande,29884416000.0,0.500394632991318
+MHA,winogrande,31457280000.0,0.4996053670086819
+MHA,winogrande,33030144000.0,0.5043409629044988
+MHA,winogrande,34603008000.0,0.5019731649565904
+MHA,winogrande,36175872000.0,0.5067087608524072
+MHA,winogrande,37748736000.0,0.4940805051302289
+MHA,winogrande,39321600000.0,0.4964483030781373
+MHA,winogrande,40894464000.0,0.5082872928176796
+MHA,winogrande,42467328000.0,0.49013417521704816
+MHA,winogrande,44040192000.0,0.4940805051302289
+MHA,winogrande,45613056000.0,0.5019731649565904
+MHA,winogrande,47185920000.0,0.5019731649565904
+MHA,winogrande,47185920000.0,0.5019731649565904
+MHA,openbookqa,1572864000.0,0.24
+MHA,openbookqa,3145728000.0,0.264
+MHA,openbookqa,4718592000.0,0.27
+MHA,openbookqa,6291456000.0,0.272
+MHA,openbookqa,7864320000.0,0.294
+MHA,openbookqa,9437184000.0,0.284
+MHA,openbookqa,11010048000.0,0.304
+MHA,openbookqa,12582912000.0,0.286
+MHA,openbookqa,14155776000.0,0.28
+MHA,openbookqa,15728640000.0,0.306
+MHA,openbookqa,17301504000.0,0.3
+MHA,openbookqa,18874368000.0,0.298
+MHA,openbookqa,20447232000.0,0.302
+MHA,openbookqa,22020096000.0,0.3
+MHA,openbookqa,23592960000.0,0.312
+MHA,openbookqa,25165824000.0,0.304
+MHA,openbookqa,26738688000.0,0.302
+MHA,openbookqa,28311552000.0,0.31
+MHA,openbookqa,29884416000.0,0.306
+MHA,openbookqa,31457280000.0,0.308
+MHA,openbookqa,33030144000.0,0.306
+MHA,openbookqa,34603008000.0,0.3
+MHA,openbookqa,36175872000.0,0.316
+MHA,openbookqa,37748736000.0,0.32
+MHA,openbookqa,39321600000.0,0.308
+MHA,openbookqa,40894464000.0,0.32
+MHA,openbookqa,42467328000.0,0.316
+MHA,openbookqa,44040192000.0,0.32
+MHA,openbookqa,45613056000.0,0.33
+MHA,openbookqa,47185920000.0,0.318
+MHA,openbookqa,47185920000.0,0.318
+GQA 4 groups,hellaswag,1572864000.0,0.26309500099581756
+GQA 4 groups,hellaswag,3145728000.0,0.2975502887870942
+GQA 4 groups,hellaswag,4718592000.0,0.3269269069906393
+GQA 4 groups,hellaswag,6291456000.0,0.3497311292571201
+GQA 4 groups,hellaswag,7864320000.0,0.36496713802031466
+GQA 4 groups,hellaswag,9437184000.0,0.38090021907986454
+GQA 4 groups,hellaswag,11010048000.0,0.3994224258115913
+GQA 4 groups,hellaswag,12582912000.0,0.40818562039434375
+GQA 4 groups,hellaswag,14155776000.0,0.4146584345747859
+GQA 4 groups,hellaswag,15728640000.0,0.4244174467237602
+GQA 4 groups,hellaswag,17301504000.0,0.43328022306313485
+GQA 4 groups,hellaswag,18874368000.0,0.4400517825134435
+GQA 4 groups,hellaswag,20447232000.0,0.4453296156144194
+GQA 4 groups,hellaswag,22020096000.0,0.4519020115514838
+GQA 4 groups,hellaswag,23592960000.0,0.4609639514041028
+GQA 4 groups,hellaswag,25165824000.0,0.4645488946425015
+GQA 4 groups,hellaswag,26738688000.0,0.4702250547699661
+GQA 4 groups,hellaswag,28311552000.0,0.4718183628759211
+GQA 4 groups,hellaswag,29884416000.0,0.477096195976897
+GQA 4 groups,hellaswag,31457280000.0,0.4814777932682733
+GQA 4 groups,hellaswag,33030144000.0,0.48466440948018324
+GQA 4 groups,hellaswag,34603008000.0,0.48685520812587135
+GQA 4 groups,hellaswag,36175872000.0,0.48994224258115915
+GQA 4 groups,hellaswag,37748736000.0,0.4963154750049791
+GQA 4 groups,hellaswag,39321600000.0,0.49591714797849035
+GQA 4 groups,hellaswag,40894464000.0,0.4985062736506672
+GQA 4 groups,hellaswag,42467328000.0,0.5007966540529775
+GQA 4 groups,hellaswag,44040192000.0,0.5006970722963553
+GQA 4 groups,hellaswag,45613056000.0,0.504282015534754
+GQA 4 groups,hellaswag,47185920000.0,0.5061740689105756
+GQA 4 groups,hellaswag,47185920000.0,0.5064728141804421
+GQA 4 groups,mmlu,1572864000.0,0.25817387119066715
+GQA 4 groups,mmlu,3145728000.0,0.26866486739979367
+GQA 4 groups,mmlu,4718592000.0,0.2812754303683095
+GQA 4 groups,mmlu,6291456000.0,0.285038311563633
+GQA 4 groups,mmlu,7864320000.0,0.2938572790136739
+GQA 4 groups,mmlu,9437184000.0,0.29513617735690134
+GQA 4 groups,mmlu,11010048000.0,0.293335083172075
+GQA 4 groups,mmlu,12582912000.0,0.30641700760387536
+GQA 4 groups,mmlu,14155776000.0,0.3060554134289484
+GQA 4 groups,mmlu,15728640000.0,0.3062895456055499
+GQA 4 groups,mmlu,17301504000.0,0.30872836661799985
+GQA 4 groups,mmlu,18874368000.0,0.3076957383403891
+GQA 4 groups,mmlu,20447232000.0,0.30909121047820975
+GQA 4 groups,mmlu,22020096000.0,0.3146701805977974
+GQA 4 groups,mmlu,23592960000.0,0.315150961665562
+GQA 4 groups,mmlu,25165824000.0,0.3171814877807358
+GQA 4 groups,mmlu,26738688000.0,0.3207778430923623
+GQA 4 groups,mmlu,28311552000.0,0.3193286717868589
+GQA 4 groups,mmlu,29884416000.0,0.32440298972136605
+GQA 4 groups,mmlu,31457280000.0,0.3237774618570344
+GQA 4 groups,mmlu,33030144000.0,0.3247659095596553
+GQA 4 groups,mmlu,34603008000.0,0.33177611671583224
+GQA 4 groups,mmlu,36175872000.0,0.3247801796318973
+GQA 4 groups,mmlu,37748736000.0,0.32446368090161676
+GQA 4 groups,mmlu,39321600000.0,0.3310436191440769
+GQA 4 groups,mmlu,40894464000.0,0.3343864297238491
+GQA 4 groups,mmlu,42467328000.0,0.3326059664944829
+GQA 4 groups,mmlu,44040192000.0,0.3313856080585158
+GQA 4 groups,mmlu,45613056000.0,0.3313425496930281
+GQA 4 groups,mmlu,47185920000.0,0.3316693261630883
+GQA 4 groups,mmlu,47185920000.0,0.33201964793742805
+GQA 4 groups,arc,1572864000.0,0.2840830086990496
+GQA 4 groups,arc,3145728000.0,0.3361702922283127
+GQA 4 groups,arc,4718592000.0,0.3655647774675078
+GQA 4 groups,arc,6291456000.0,0.3911886211374266
+GQA 4 groups,arc,7864320000.0,0.3938725997173096
+GQA 4 groups,arc,9437184000.0,0.4031548419347054
+GQA 4 groups,arc,11010048000.0,0.4127164707369486
+GQA 4 groups,arc,12582912000.0,0.4232900679146413
+GQA 4 groups,arc,14155776000.0,0.435336729065398
+GQA 4 groups,arc,15728640000.0,0.42957447053010195
+GQA 4 groups,arc,17301504000.0,0.43953973753461806
+GQA 4 groups,arc,18874368000.0,0.4464726904999943
+GQA 4 groups,arc,20447232000.0,0.4528432792084669
+GQA 4 groups,arc,22020096000.0,0.4553570402546512
+GQA 4 groups,arc,23592960000.0,0.4527915675526597
+GQA 4 groups,arc,25165824000.0,0.4646220452534445
+GQA 4 groups,arc,26738688000.0,0.46610660070557686
+GQA 4 groups,arc,28311552000.0,0.46930913227841553
+GQA 4 groups,arc,29884416000.0,0.46401946081980217
+GQA 4 groups,arc,31457280000.0,0.47313148550349915
+GQA 4 groups,arc,33030144000.0,0.4686663276680342
+GQA 4 groups,arc,34603008000.0,0.481122372760598
+GQA 4 groups,arc,36175872000.0,0.4753945886625067
+GQA 4 groups,arc,37748736000.0,0.4790352328748233
+GQA 4 groups,arc,39321600000.0,0.47289231909539076
+GQA 4 groups,arc,40894464000.0,0.4849217430275451
+GQA 4 groups,arc,42467328000.0,0.489427121039749
+GQA 4 groups,arc,44040192000.0,0.4883289665712874
+GQA 4 groups,arc,45613056000.0,0.48072448029785914
+GQA 4 groups,arc,47185920000.0,0.4843651245101757
+GQA 4 groups,arc,47185920000.0,0.4830967525080153
+GQA 4 groups,piqa,1572864000.0,0.5631120783460283
+GQA 4 groups,piqa,3145728000.0,0.5979325353645266
+GQA 4 groups,piqa,4718592000.0,0.6245919477693145
+GQA 4 groups,piqa,6291456000.0,0.6360174102285092
+GQA 4 groups,piqa,7864320000.0,0.6441784548422198
+GQA 4 groups,piqa,9437184000.0,0.6605005440696409
+GQA 4 groups,piqa,11010048000.0,0.6659412404787813
+GQA 4 groups,piqa,12582912000.0,0.6746463547334058
+GQA 4 groups,piqa,14155776000.0,0.676822633297062
+GQA 4 groups,piqa,15728640000.0,0.6828073993471164
+GQA 4 groups,piqa,17301504000.0,0.6860718171926007
+GQA 4 groups,piqa,18874368000.0,0.6887921653971708
+GQA 4 groups,piqa,20447232000.0,0.6887921653971708
+GQA 4 groups,piqa,22020096000.0,0.6822633297062024
+GQA 4 groups,piqa,23592960000.0,0.6871599564744287
+GQA 4 groups,piqa,25165824000.0,0.6920565832426551
+GQA 4 groups,piqa,26738688000.0,0.7002176278563657
+GQA 4 groups,piqa,28311552000.0,0.6969532100108814
+GQA 4 groups,piqa,29884416000.0,0.6974972796517954
+GQA 4 groups,piqa,31457280000.0,0.6964091403699674
+GQA 4 groups,piqa,33030144000.0,0.705114254624592
+GQA 4 groups,piqa,34603008000.0,0.705114254624592
+GQA 4 groups,piqa,36175872000.0,0.7040261153427638
+GQA 4 groups,piqa,37748736000.0,0.7002176278563657
+GQA 4 groups,piqa,39321600000.0,0.7002176278563657
+GQA 4 groups,piqa,40894464000.0,0.7029379760609358
+GQA 4 groups,piqa,42467328000.0,0.7110990206746464
+GQA 4 groups,piqa,44040192000.0,0.7159956474428727
+GQA 4 groups,piqa,45613056000.0,0.7132752992383025
+GQA 4 groups,piqa,47185920000.0,0.7094668117519043
+GQA 4 groups,piqa,47185920000.0,0.7105549510337323
+GQA 4 groups,commonsenseqa,1572864000.0,0.2571662571662572
+GQA 4 groups,commonsenseqa,3145728000.0,0.29565929565929566
+GQA 4 groups,commonsenseqa,4718592000.0,0.32432432432432434
+GQA 4 groups,commonsenseqa,6291456000.0,0.32186732186732187
+GQA 4 groups,commonsenseqa,7864320000.0,0.32678132678132676
+GQA 4 groups,commonsenseqa,9437184000.0,0.32596232596232594
+GQA 4 groups,commonsenseqa,11010048000.0,0.3325143325143325
+GQA 4 groups,commonsenseqa,12582912000.0,0.3751023751023751
+GQA 4 groups,commonsenseqa,14155776000.0,0.37346437346437344
+GQA 4 groups,commonsenseqa,15728640000.0,0.36855036855036855
+GQA 4 groups,commonsenseqa,17301504000.0,0.38656838656838655
+GQA 4 groups,commonsenseqa,18874368000.0,0.3783783783783784
+GQA 4 groups,commonsenseqa,20447232000.0,0.3595413595413595
+GQA 4 groups,commonsenseqa,22020096000.0,0.37182637182637185
+GQA 4 groups,commonsenseqa,23592960000.0,0.3726453726453726
+GQA 4 groups,commonsenseqa,25165824000.0,0.36936936936936937
+GQA 4 groups,commonsenseqa,26738688000.0,0.3775593775593776
+GQA 4 groups,commonsenseqa,28311552000.0,0.3824733824733825
+GQA 4 groups,commonsenseqa,29884416000.0,0.389025389025389
+GQA 4 groups,commonsenseqa,31457280000.0,0.4013104013104013
+GQA 4 groups,commonsenseqa,33030144000.0,0.3841113841113841
+GQA 4 groups,commonsenseqa,34603008000.0,0.39803439803439805
+GQA 4 groups,commonsenseqa,36175872000.0,0.38656838656838655
+GQA 4 groups,commonsenseqa,37748736000.0,0.38738738738738737
+GQA 4 groups,commonsenseqa,39321600000.0,0.3759213759213759
+GQA 4 groups,commonsenseqa,40894464000.0,0.3988533988533989
+GQA 4 groups,commonsenseqa,42467328000.0,0.389025389025389
+GQA 4 groups,commonsenseqa,44040192000.0,0.4021294021294021
+GQA 4 groups,commonsenseqa,45613056000.0,0.38738738738738737
+GQA 4 groups,commonsenseqa,47185920000.0,0.3841113841113841
+GQA 4 groups,commonsenseqa,47185920000.0,0.3841113841113841
+GQA 4 groups,winogrande,1572864000.0,0.4925019731649566
+GQA 4 groups,winogrande,3145728000.0,0.48382004735595896
+GQA 4 groups,winogrande,4718592000.0,0.5011838989739542
+GQA 4 groups,winogrande,6291456000.0,0.4909234411996843
+GQA 4 groups,winogrande,7864320000.0,0.4925019731649566
+GQA 4 groups,winogrande,9437184000.0,0.48382004735595896
+GQA 4 groups,winogrande,11010048000.0,0.4925019731649566
+GQA 4 groups,winogrande,12582912000.0,0.48855564325177586
+GQA 4 groups,winogrande,14155776000.0,0.4940805051302289
+GQA 4 groups,winogrande,15728640000.0,0.5019731649565904
+GQA 4 groups,winogrande,17301504000.0,0.49329123914759276
+GQA 4 groups,winogrande,18874368000.0,0.5027624309392266
+GQA 4 groups,winogrande,20447232000.0,0.4988161010260458
+GQA 4 groups,winogrande,22020096000.0,0.5114443567482242
+GQA 4 groups,winogrande,23592960000.0,0.4996053670086819
+GQA 4 groups,winogrande,25165824000.0,0.4956590370955012
+GQA 4 groups,winogrande,26738688000.0,0.500394632991318
+GQA 4 groups,winogrande,28311552000.0,0.5035516969218626
+GQA 4 groups,winogrande,29884416000.0,0.4940805051302289
+GQA 4 groups,winogrande,31457280000.0,0.5019731649565904
+GQA 4 groups,winogrande,33030144000.0,0.4925019731649566
+GQA 4 groups,winogrande,34603008000.0,0.4980268350434096
+GQA 4 groups,winogrande,36175872000.0,0.4972375690607735
+GQA 4 groups,winogrande,37748736000.0,0.4988161010260458
+GQA 4 groups,winogrande,39321600000.0,0.4980268350434096
+GQA 4 groups,winogrande,40894464000.0,0.4988161010260458
+GQA 4 groups,winogrande,42467328000.0,0.49329123914759276
+GQA 4 groups,winogrande,44040192000.0,0.48855564325177586
+GQA 4 groups,winogrande,45613056000.0,0.494869771112865
+GQA 4 groups,winogrande,47185920000.0,0.4988161010260458
+GQA 4 groups,winogrande,47185920000.0,0.4980268350434096
+GQA 4 groups,openbookqa,1572864000.0,0.254
+GQA 4 groups,openbookqa,3145728000.0,0.236
+GQA 4 groups,openbookqa,4718592000.0,0.268
+GQA 4 groups,openbookqa,6291456000.0,0.272
+GQA 4 groups,openbookqa,7864320000.0,0.272
+GQA 4 groups,openbookqa,9437184000.0,0.282
+GQA 4 groups,openbookqa,11010048000.0,0.314
+GQA 4 groups,openbookqa,12582912000.0,0.284
+GQA 4 groups,openbookqa,14155776000.0,0.282
+GQA 4 groups,openbookqa,15728640000.0,0.292
+GQA 4 groups,openbookqa,17301504000.0,0.298
+GQA 4 groups,openbookqa,18874368000.0,0.288
+GQA 4 groups,openbookqa,20447232000.0,0.306
+GQA 4 groups,openbookqa,22020096000.0,0.288
+GQA 4 groups,openbookqa,23592960000.0,0.308
+GQA 4 groups,openbookqa,25165824000.0,0.328
+GQA 4 groups,openbookqa,26738688000.0,0.312
+GQA 4 groups,openbookqa,28311552000.0,0.312
+GQA 4 groups,openbookqa,29884416000.0,0.312
+GQA 4 groups,openbookqa,31457280000.0,0.314
+GQA 4 groups,openbookqa,33030144000.0,0.324
+GQA 4 groups,openbookqa,34603008000.0,0.318
+GQA 4 groups,openbookqa,36175872000.0,0.308
+GQA 4 groups,openbookqa,37748736000.0,0.312
+GQA 4 groups,openbookqa,39321600000.0,0.328
+GQA 4 groups,openbookqa,40894464000.0,0.328
+GQA 4 groups,openbookqa,42467328000.0,0.33
+GQA 4 groups,openbookqa,44040192000.0,0.322
+GQA 4 groups,openbookqa,45613056000.0,0.31
+GQA 4 groups,openbookqa,47185920000.0,0.322
+GQA 4 groups,openbookqa,47185920000.0,0.314
+MQA,hellaswag,1572864000.0,0.26110336586337385
+MQA,hellaswag,3145728000.0,0.2898824935271858
+MQA,hellaswag,4718592000.0,0.31866162119099783
+MQA,hellaswag,6291456000.0,0.3405696076478789
+MQA,hellaswag,7864320000.0,0.35839474208325034
+MQA,hellaswag,9437184000.0,0.37333200557657836
+MQA,hellaswag,11010048000.0,0.3906592312288389
+MQA,hellaswag,12582912000.0,0.3975303724357698
+MQA,hellaswag,14155776000.0,0.40888269269069905
+MQA,hellaswag,15728640000.0,0.41545508862776337
+MQA,hellaswag,17301504000.0,0.4222266480780721
+MQA,hellaswag,18874368000.0,0.42471619199362676
+MQA,hellaswag,20447232000.0,0.4284007169886477
+MQA,hellaswag,22020096000.0,0.43397729535949015
+MQA,hellaswag,23592960000.0,0.4431388169687313
+MQA,hellaswag,25165824000.0,0.448814977096196
+MQA,hellaswag,26738688000.0,0.451503684524995
+MQA,hellaswag,28311552000.0,0.45578570005974905
+MQA,hellaswag,29884416000.0,0.4627564230233021
+MQA,hellaswag,31457280000.0,0.46683927504481176
+MQA,hellaswag,33030144000.0,0.4695279824736108
+MQA,hellaswag,34603008000.0,0.4731129257120096
+MQA,hellaswag,36175872000.0,0.4759012148974308
+MQA,hellaswag,37748736000.0,0.47649870543716394
+MQA,hellaswag,39321600000.0,0.47829117705636326
+MQA,hellaswag,40894464000.0,0.48028281218880703
+MQA,hellaswag,42467328000.0,0.4810794662417845
+MQA,hellaswag,44040192000.0,0.4836685919139614
+MQA,hellaswag,45613056000.0,0.48595897231627166
+MQA,hellaswag,47185920000.0,0.4871539533957379
+MQA,hellaswag,47185920000.0,0.4871539533957379
+MQA,mmlu,1572864000.0,0.2543627197648624
+MQA,mmlu,3145728000.0,0.2681902386009931
+MQA,mmlu,4718592000.0,0.27721373975226216
+MQA,mmlu,6291456000.0,0.2843085595466949
+MQA,mmlu,7864320000.0,0.28932350587064265
+MQA,mmlu,9437184000.0,0.29144249967506186
+MQA,mmlu,11010048000.0,0.2959225754896006
+MQA,mmlu,12582912000.0,0.295126698843318
+MQA,mmlu,14155776000.0,0.2971442266521409
+MQA,mmlu,15728640000.0,0.30235278230680646
+MQA,mmlu,17301504000.0,0.3019403965700273
+MQA,mmlu,18874368000.0,0.301417119644089
+MQA,mmlu,20447232000.0,0.30707874634000326
+MQA,mmlu,22020096000.0,0.30769649373713626
+MQA,mmlu,23592960000.0,0.3062950734734772
+MQA,mmlu,25165824000.0,0.30896455785273996
+MQA,mmlu,26738688000.0,0.31141116214089487
+MQA,mmlu,28311552000.0,0.31482284600059135
+MQA,mmlu,29884416000.0,0.32057271598555037
+MQA,mmlu,31457280000.0,0.318793334037572
+MQA,mmlu,33030144000.0,0.3199069407624995
+MQA,mmlu,34603008000.0,0.32339578170084887
+MQA,mmlu,36175872000.0,0.32109429952337987
+MQA,mmlu,37748736000.0,0.3215193776213223
+MQA,mmlu,39321600000.0,0.32339743197770454
+MQA,mmlu,40894464000.0,0.3268698924129341
+MQA,mmlu,42467328000.0,0.3265744392425102
+MQA,mmlu,44040192000.0,0.32592224844954487
+MQA,mmlu,45613056000.0,0.3227556242964135
+MQA,mmlu,47185920000.0,0.32461250872284597
+MQA,mmlu,47185920000.0,0.32461250872284597
+MQA,arc,1572864000.0,0.28476028774663587
+MQA,arc,3145728000.0,0.3291683903885269
+MQA,arc,4718592000.0,0.3574934498569311
+MQA,arc,6291456000.0,0.372316021420117
+MQA,arc,7864320000.0,0.38247448891646846
+MQA,arc,9437184000.0,0.39697745371806803
+MQA,arc,11010048000.0,0.4102924868709852
+MQA,arc,12582912000.0,0.4219699842566737
+MQA,arc,14155776000.0,0.43232739798439457
+MQA,arc,15728640000.0,0.43212845175302517
+MQA,arc,17301504000.0,0.43548970938049436
+MQA,arc,18874368000.0,0.4354609806828237
+MQA,arc,20447232000.0,0.44224669907263764
+MQA,arc,22020096000.0,0.44269055745164965
+MQA,arc,23592960000.0,0.4386232920789235
+MQA,arc,25165824000.0,0.4505054814355156
+MQA,arc,26738688000.0,0.4470580377150343
+MQA,arc,28311552000.0,0.4494590386228612
+MQA,arc,29884416000.0,0.44855983038576897
+MQA,arc,31457280000.0,0.458701060663518
+MQA,arc,33030144000.0,0.45957154020293955
+MQA,arc,34603008000.0,0.47080517920961606
+MQA,arc,36175872000.0,0.4670805035566128
+MQA,arc,37748736000.0,0.46897444295055213
+MQA,arc,39321600000.0,0.46469673986738835
+MQA,arc,40894464000.0,0.47148245825720225
+MQA,arc,42467328000.0,0.46493590627549675
+MQA,arc,44040192000.0,0.4714767125176681
+MQA,arc,45613056000.0,0.4769768216867193
+MQA,arc,47185920000.0,0.4680867261925282
+MQA,arc,47185920000.0,0.4680867261925282
+MQA,piqa,1572864000.0,0.5505984766050055
+MQA,piqa,3145728000.0,0.5968443960826986
+MQA,piqa,4718592000.0,0.6158868335146899
+MQA,piqa,6291456000.0,0.6218715995647442
+MQA,piqa,7864320000.0,0.6517954298150164
+MQA,piqa,9437184000.0,0.6648531011969532
+MQA,piqa,11010048000.0,0.6632208922742111
+MQA,piqa,12582912000.0,0.6637649619151251
+MQA,piqa,14155776000.0,0.6708378672470077
+MQA,piqa,15728640000.0,0.6719260065288357
+MQA,piqa,17301504000.0,0.6800870511425462
+MQA,piqa,18874368000.0,0.6702937976060935
+MQA,piqa,20447232000.0,0.6789989118607181
+MQA,piqa,22020096000.0,0.6833514689880305
+MQA,piqa,23592960000.0,0.6898803046789989
+MQA,piqa,25165824000.0,0.6877040261153428
+MQA,piqa,26738688000.0,0.6985854189336235
+MQA,piqa,28311552000.0,0.6931447225244831
+MQA,piqa,29884416000.0,0.6926006528835691
+MQA,piqa,31457280000.0,0.6964091403699674
+MQA,piqa,33030144000.0,0.6996735582154516
+MQA,piqa,34603008000.0,0.6985854189336235
+MQA,piqa,36175872000.0,0.6996735582154516
+MQA,piqa,37748736000.0,0.7002176278563657
+MQA,piqa,39321600000.0,0.7002176278563657
+MQA,piqa,40894464000.0,0.7007616974972797
+MQA,piqa,42467328000.0,0.6974972796517954
+MQA,piqa,44040192000.0,0.7018498367791077
+MQA,piqa,45613056000.0,0.7013057671381937
+MQA,piqa,47185920000.0,0.7013057671381937
+MQA,piqa,47185920000.0,0.7013057671381937
+MQA,commonsenseqa,1572864000.0,0.24651924651924653
+MQA,commonsenseqa,3145728000.0,0.2784602784602785
+MQA,commonsenseqa,4718592000.0,0.3194103194103194
+MQA,commonsenseqa,6291456000.0,0.3316953316953317
+MQA,commonsenseqa,7864320000.0,0.33742833742833744
+MQA,commonsenseqa,9437184000.0,0.34807534807534807
+MQA,commonsenseqa,11010048000.0,0.3497133497133497
+MQA,commonsenseqa,12582912000.0,0.37346437346437344
+MQA,commonsenseqa,14155776000.0,0.3628173628173628
+MQA,commonsenseqa,15728640000.0,0.3701883701883702
+MQA,commonsenseqa,17301504000.0,0.36609336609336607
+MQA,commonsenseqa,18874368000.0,0.36855036855036855
+MQA,commonsenseqa,20447232000.0,0.3701883701883702
+MQA,commonsenseqa,22020096000.0,0.36936936936936937
+MQA,commonsenseqa,23592960000.0,0.3824733824733825
+MQA,commonsenseqa,25165824000.0,0.3652743652743653
+MQA,commonsenseqa,26738688000.0,0.3759213759213759
+MQA,commonsenseqa,28311552000.0,0.3783783783783784
+MQA,commonsenseqa,29884416000.0,0.37919737919737917
+MQA,commonsenseqa,31457280000.0,0.3906633906633907
+MQA,commonsenseqa,33030144000.0,0.3914823914823915
+MQA,commonsenseqa,34603008000.0,0.3972153972153972
+MQA,commonsenseqa,36175872000.0,0.38984438984438985
+MQA,commonsenseqa,37748736000.0,0.3914823914823915
+MQA,commonsenseqa,39321600000.0,0.39803439803439805
+MQA,commonsenseqa,40894464000.0,0.3955773955773956
+MQA,commonsenseqa,42467328000.0,0.39967239967239965
+MQA,commonsenseqa,44040192000.0,0.4021294021294021
+MQA,commonsenseqa,45613056000.0,0.3906633906633907
+MQA,commonsenseqa,47185920000.0,0.39475839475839475
+MQA,commonsenseqa,47185920000.0,0.39475839475839475
+MQA,winogrande,1572864000.0,0.47671665351223363
+MQA,winogrande,3145728000.0,0.4648776637726914
+MQA,winogrande,4718592000.0,0.5019731649565904
+MQA,winogrande,6291456000.0,0.500394632991318
+MQA,winogrande,7864320000.0,0.49329123914759276
+MQA,winogrande,9437184000.0,0.5059194948697711
+MQA,winogrande,11010048000.0,0.48697711128650356
+MQA,winogrande,12582912000.0,0.49171270718232046
+MQA,winogrande,14155776000.0,0.48382004735595896
+MQA,winogrande,15728640000.0,0.494869771112865
+MQA,winogrande,17301504000.0,0.4925019731649566
+MQA,winogrande,18874368000.0,0.500394632991318
+MQA,winogrande,20447232000.0,0.5011838989739542
+MQA,winogrande,22020096000.0,0.5090765588003157
+MQA,winogrande,23592960000.0,0.5074980268350434
+MQA,winogrande,25165824000.0,0.5059194948697711
+MQA,winogrande,26738688000.0,0.500394632991318
+MQA,winogrande,28311552000.0,0.5011838989739542
+MQA,winogrande,29884416000.0,0.489344909234412
+MQA,winogrande,31457280000.0,0.5059194948697711
+MQA,winogrande,33030144000.0,0.5011838989739542
+MQA,winogrande,34603008000.0,0.5027624309392266
+MQA,winogrande,36175872000.0,0.5043409629044988
+MQA,winogrande,37748736000.0,0.5090765588003157
+MQA,winogrande,39321600000.0,0.4964483030781373
+MQA,winogrande,40894464000.0,0.4996053670086819
+MQA,winogrande,42467328000.0,0.500394632991318
+MQA,winogrande,44040192000.0,0.5067087608524072
+MQA,winogrande,45613056000.0,0.4988161010260458
+MQA,winogrande,47185920000.0,0.4996053670086819
+MQA,winogrande,47185920000.0,0.4996053670086819
+MQA,openbookqa,1572864000.0,0.23
+MQA,openbookqa,3145728000.0,0.242
+MQA,openbookqa,4718592000.0,0.268
+MQA,openbookqa,6291456000.0,0.28
+MQA,openbookqa,7864320000.0,0.282
+MQA,openbookqa,9437184000.0,0.292
+MQA,openbookqa,11010048000.0,0.294
+MQA,openbookqa,12582912000.0,0.292
+MQA,openbookqa,14155776000.0,0.3
+MQA,openbookqa,15728640000.0,0.306
+MQA,openbookqa,17301504000.0,0.31
+MQA,openbookqa,18874368000.0,0.308
+MQA,openbookqa,20447232000.0,0.308
+MQA,openbookqa,22020096000.0,0.312
+MQA,openbookqa,23592960000.0,0.306
+MQA,openbookqa,25165824000.0,0.302
+MQA,openbookqa,26738688000.0,0.306
+MQA,openbookqa,28311552000.0,0.322
+MQA,openbookqa,29884416000.0,0.318
+MQA,openbookqa,31457280000.0,0.33
+MQA,openbookqa,33030144000.0,0.316
+MQA,openbookqa,34603008000.0,0.33
+MQA,openbookqa,36175872000.0,0.332
+MQA,openbookqa,37748736000.0,0.31
+MQA,openbookqa,39321600000.0,0.322
+MQA,openbookqa,40894464000.0,0.326
+MQA,openbookqa,42467328000.0,0.326
+MQA,openbookqa,44040192000.0,0.322
+MQA,openbookqa,45613056000.0,0.324
+MQA,openbookqa,47185920000.0,0.332
+MQA,openbookqa,47185920000.0,0.332
+GQA 2 groups,hellaswag,1572864000.0,0.2638916550487951
+GQA 2 groups,hellaswag,3145728000.0,0.3005377414857598
+GQA 2 groups,hellaswag,4718592000.0,0.3291177056363274
+GQA 2 groups,hellaswag,6291456000.0,0.34863572993427605
+GQA 2 groups,hellaswag,7864320000.0,0.3703445528779128
+GQA 2 groups,hellaswag,9437184000.0,0.3855805616411073
+GQA 2 groups,hellaswag,11010048000.0,0.40111531567416847
+GQA 2 groups,hellaswag,12582912000.0,0.410973909579765
+GQA 2 groups,hellaswag,14155776000.0,0.4221270663214499
+GQA 2 groups,hellaswag,15728640000.0,0.42850029874526985
+GQA 2 groups,hellaswag,17301504000.0,0.43835889265086636
+GQA 2 groups,hellaswag,18874368000.0,0.4425413264289982
+GQA 2 groups,hellaswag,20447232000.0,0.4470225054769966
+GQA 2 groups,hellaswag,22020096000.0,0.45269866560446126
+GQA 2 groups,hellaswag,23592960000.0,0.4596693885680143
+GQA 2 groups,hellaswag,25165824000.0,0.4666401115315674
+GQA 2 groups,hellaswag,26738688000.0,0.4682334196375224
+GQA 2 groups,hellaswag,28311552000.0,0.472814180442143
+GQA 2 groups,hellaswag,29884416000.0,0.47679745070703045
+GQA 2 groups,hellaswag,31457280000.0,0.48247361083449514
+GQA 2 groups,hellaswag,33030144000.0,0.4827723561043617
+GQA 2 groups,hellaswag,34603008000.0,0.4878510256920932
+GQA 2 groups,hellaswag,36175872000.0,0.4901414060944035
+GQA 2 groups,hellaswag,37748736000.0,0.49283011352320255
+GQA 2 groups,hellaswag,39321600000.0,0.49482174865564627
+GQA 2 groups,hellaswag,40894464000.0,0.4954192391953794
+GQA 2 groups,hellaswag,42467328000.0,0.4989046006771559
+GQA 2 groups,hellaswag,44040192000.0,0.5008962358095996
+GQA 2 groups,hellaswag,45613056000.0,0.500497908783111
+GQA 2 groups,hellaswag,47185920000.0,0.5035849432383988
+GQA 2 groups,hellaswag,47185920000.0,0.5036845249950209
+GQA 2 groups,mmlu,1572864000.0,0.25893815901560957
+GQA 2 groups,mmlu,3145728000.0,0.266950838336094
+GQA 2 groups,mmlu,4718592000.0,0.2782766769711504
+GQA 2 groups,mmlu,6291456000.0,0.2806269796728518
+GQA 2 groups,mmlu,7864320000.0,0.28977318002513425
+GQA 2 groups,mmlu,9437184000.0,0.2952406112618376
+GQA 2 groups,mmlu,11010048000.0,0.2997183871428359
+GQA 2 groups,mmlu,12582912000.0,0.29655800059744747
+GQA 2 groups,mmlu,14155776000.0,0.30496467062704363
+GQA 2 groups,mmlu,15728640000.0,0.2996148691666964
+GQA 2 groups,mmlu,17301504000.0,0.3113110420783124
+GQA 2 groups,mmlu,18874368000.0,0.3132468492529825
+GQA 2 groups,mmlu,20447232000.0,0.3151472361472257
+GQA 2 groups,mmlu,22020096000.0,0.3129632286933616
+GQA 2 groups,mmlu,23592960000.0,0.3170194969721171
+GQA 2 groups,mmlu,25165824000.0,0.3165538196067337
+GQA 2 groups,mmlu,26738688000.0,0.3167847202070646
+GQA 2 groups,mmlu,28311552000.0,0.3231019852425221
+GQA 2 groups,mmlu,29884416000.0,0.3225677506747409
+GQA 2 groups,mmlu,31457280000.0,0.32550547783751416
+GQA 2 groups,mmlu,33030144000.0,0.3242943906465741
+GQA 2 groups,mmlu,34603008000.0,0.3228672462593321
+GQA 2 groups,mmlu,36175872000.0,0.3231552782301944
+GQA 2 groups,mmlu,37748736000.0,0.3282547471790177
+GQA 2 groups,mmlu,39321600000.0,0.3278489908690609
+GQA 2 groups,mmlu,40894464000.0,0.33286469192239854
+GQA 2 groups,mmlu,42467328000.0,0.33491847939598235
+GQA 2 groups,mmlu,44040192000.0,0.3283555078544681
+GQA 2 groups,mmlu,45613056000.0,0.3307858249400203
+GQA 2 groups,mmlu,47185920000.0,0.32988578157607207
+GQA 2 groups,mmlu,47185920000.0,0.32962608134765653
+GQA 2 groups,arc,1572864000.0,0.2872223371370129
+GQA 2 groups,arc,3145728000.0,0.3385080900012641
+GQA 2 groups,arc,4718592000.0,0.36550732007216646
+GQA 2 groups,arc,6291456000.0,0.38583574654393765
+GQA 2 groups,arc,7864320000.0,0.3887646372714632
+GQA 2 groups,arc,9437184000.0,0.41177129658358325
+GQA 2 groups,arc,11010048000.0,0.4042415049240988
+GQA 2 groups,arc,12582912000.0,0.41798890497695956
+GQA 2 groups,arc,14155776000.0,0.433999408188828
+GQA 2 groups,arc,15728640000.0,0.4323503809425311
+GQA 2 groups,arc,17301504000.0,0.44124047643672215
+GQA 2 groups,arc,18874368000.0,0.4554087519104584
+GQA 2 groups,arc,20447232000.0,0.44249161122028013
+GQA 2 groups,arc,22020096000.0,0.43615549694901234
+GQA 2 groups,arc,23592960000.0,0.43779877845577503
+GQA 2 groups,arc,25165824000.0,0.45561344388136193
+GQA 2 groups,arc,26738688000.0,0.45742694292182345
+GQA 2 groups,arc,28311552000.0,0.45663690373587984
+GQA 2 groups,arc,29884416000.0,0.4613239907608508
+GQA 2 groups,arc,31457280000.0,0.47271061008262377
+GQA 2 groups,arc,33030144000.0,0.47610634214729775
+GQA 2 groups,arc,34603008000.0,0.4750484078555751
+GQA 2 groups,arc,36175872000.0,0.4697472449178933
+GQA 2 groups,arc,37748736000.0,0.4767146723204744
+GQA 2 groups,arc,39321600000.0,0.4710041254409855
+GQA 2 groups,arc,40894464000.0,0.4756107721124786
+GQA 2 groups,arc,42467328000.0,0.4856334965123361
+GQA 2 groups,arc,44040192000.0,0.48626480964364927
+GQA 2 groups,arc,45613056000.0,0.483751048597465
+GQA 2 groups,arc,47185920000.0,0.4794963284724377
+GQA 2 groups,arc,47185920000.0,0.47991720389331316
+GQA 2 groups,piqa,1572864000.0,0.558759521218716
+GQA 2 groups,piqa,3145728000.0,0.5957562568008705
+GQA 2 groups,piqa,4718592000.0,0.6169749727965179
+GQA 2 groups,piqa,6291456000.0,0.6305767138193689
+GQA 2 groups,piqa,7864320000.0,0.6577801958650707
+GQA 2 groups,piqa,9437184000.0,0.6643090315560392
+GQA 2 groups,piqa,11010048000.0,0.6806311207834603
+GQA 2 groups,piqa,12582912000.0,0.676822633297062
+GQA 2 groups,piqa,14155776000.0,0.6719260065288357
+GQA 2 groups,piqa,15728640000.0,0.6735582154515778
+GQA 2 groups,piqa,17301504000.0,0.6741022850924918
+GQA 2 groups,piqa,18874368000.0,0.6887921653971708
+GQA 2 groups,piqa,20447232000.0,0.6969532100108814
+GQA 2 groups,piqa,22020096000.0,0.6931447225244831
+GQA 2 groups,piqa,23592960000.0,0.6866158868335147
+GQA 2 groups,piqa,25165824000.0,0.6893362350380848
+GQA 2 groups,piqa,26738688000.0,0.7013057671381937
+GQA 2 groups,piqa,28311552000.0,0.6974972796517954
+GQA 2 groups,piqa,29884416000.0,0.705658324265506
+GQA 2 groups,piqa,31457280000.0,0.704570184983678
+GQA 2 groups,piqa,33030144000.0,0.7121871599564744
+GQA 2 groups,piqa,34603008000.0,0.7023939064200218
+GQA 2 groups,piqa,36175872000.0,0.704570184983678
+GQA 2 groups,piqa,37748736000.0,0.7083786724700761
+GQA 2 groups,piqa,39321600000.0,0.7094668117519043
+GQA 2 groups,piqa,40894464000.0,0.7149075081610446
+GQA 2 groups,piqa,42467328000.0,0.7154515778019587
+GQA 2 groups,piqa,44040192000.0,0.7094668117519043
+GQA 2 groups,piqa,45613056000.0,0.7149075081610446
+GQA 2 groups,piqa,47185920000.0,0.7127312295973884
+GQA 2 groups,piqa,47185920000.0,0.7127312295973884
+GQA 2 groups,commonsenseqa,1572864000.0,0.25143325143325146
+GQA 2 groups,commonsenseqa,3145728000.0,0.29811629811629814
+GQA 2 groups,commonsenseqa,4718592000.0,0.3185913185913186
+GQA 2 groups,commonsenseqa,6291456000.0,0.34316134316134317
+GQA 2 groups,commonsenseqa,7864320000.0,0.3357903357903358
+GQA 2 groups,commonsenseqa,9437184000.0,0.35544635544635544
+GQA 2 groups,commonsenseqa,11010048000.0,0.3415233415233415
+GQA 2 groups,commonsenseqa,12582912000.0,0.3669123669123669
+GQA 2 groups,commonsenseqa,14155776000.0,0.36936936936936937
+GQA 2 groups,commonsenseqa,15728640000.0,0.3652743652743653
+GQA 2 groups,commonsenseqa,17301504000.0,0.3726453726453726
+GQA 2 groups,commonsenseqa,18874368000.0,0.37346437346437344
+GQA 2 groups,commonsenseqa,20447232000.0,0.35626535626535627
+GQA 2 groups,commonsenseqa,22020096000.0,0.38001638001638
+GQA 2 groups,commonsenseqa,23592960000.0,0.37428337428337427
+GQA 2 groups,commonsenseqa,25165824000.0,0.3701883701883702
+GQA 2 groups,commonsenseqa,26738688000.0,0.3939393939393939
+GQA 2 groups,commonsenseqa,28311552000.0,0.37919737919737917
+GQA 2 groups,commonsenseqa,29884416000.0,0.38493038493038495
+GQA 2 groups,commonsenseqa,31457280000.0,0.39803439803439805
+GQA 2 groups,commonsenseqa,33030144000.0,0.3914823914823915
+GQA 2 groups,commonsenseqa,34603008000.0,0.4021294021294021
+GQA 2 groups,commonsenseqa,36175872000.0,0.3759213759213759
+GQA 2 groups,commonsenseqa,37748736000.0,0.3939393939393939
+GQA 2 groups,commonsenseqa,39321600000.0,0.38656838656838655
+GQA 2 groups,commonsenseqa,40894464000.0,0.39475839475839475
+GQA 2 groups,commonsenseqa,42467328000.0,0.3857493857493858
+GQA 2 groups,commonsenseqa,44040192000.0,0.3931203931203931
+GQA 2 groups,commonsenseqa,45613056000.0,0.39967239967239965
+GQA 2 groups,commonsenseqa,47185920000.0,0.39475839475839475
+GQA 2 groups,commonsenseqa,47185920000.0,0.39475839475839475
+GQA 2 groups,winogrande,1572864000.0,0.49329123914759276
+GQA 2 groups,winogrande,3145728000.0,0.500394632991318
+GQA 2 groups,winogrande,4718592000.0,0.5019731649565904
+GQA 2 groups,winogrande,6291456000.0,0.48303078137332284
+GQA 2 groups,winogrande,7864320000.0,0.4925019731649566
+GQA 2 groups,winogrande,9437184000.0,0.5067087608524072
+GQA 2 groups,winogrande,11010048000.0,0.4988161010260458
+GQA 2 groups,winogrande,12582912000.0,0.5011838989739542
+GQA 2 groups,winogrande,14155776000.0,0.4925019731649566
+GQA 2 groups,winogrande,15728640000.0,0.4980268350434096
+GQA 2 groups,winogrande,17301504000.0,0.49013417521704816
+GQA 2 groups,winogrande,18874368000.0,0.489344909234412
+GQA 2 groups,winogrande,20447232000.0,0.49329123914759276
+GQA 2 groups,winogrande,22020096000.0,0.5011838989739542
+GQA 2 groups,winogrande,23592960000.0,0.4846093133385951
+GQA 2 groups,winogrande,25165824000.0,0.4925019731649566
+GQA 2 groups,winogrande,26738688000.0,0.48855564325177586
+GQA 2 groups,winogrande,28311552000.0,0.5059194948697711
+GQA 2 groups,winogrande,29884416000.0,0.4940805051302289
+GQA 2 groups,winogrande,31457280000.0,0.5067087608524072
+GQA 2 groups,winogrande,33030144000.0,0.5153906866614049
+GQA 2 groups,winogrande,34603008000.0,0.5035516969218626
+GQA 2 groups,winogrande,36175872000.0,0.5074980268350434
+GQA 2 groups,winogrande,37748736000.0,0.5059194948697711
+GQA 2 groups,winogrande,39321600000.0,0.505130228887135
+GQA 2 groups,winogrande,40894464000.0,0.4988161010260458
+GQA 2 groups,winogrande,42467328000.0,0.4996053670086819
+GQA 2 groups,winogrande,44040192000.0,0.494869771112865
+GQA 2 groups,winogrande,45613056000.0,0.505130228887135
+GQA 2 groups,winogrande,47185920000.0,0.5011838989739542
+GQA 2 groups,winogrande,47185920000.0,0.5011838989739542
+GQA 2 groups,openbookqa,1572864000.0,0.266
+GQA 2 groups,openbookqa,3145728000.0,0.264
+GQA 2 groups,openbookqa,4718592000.0,0.28
+GQA 2 groups,openbookqa,6291456000.0,0.282
+GQA 2 groups,openbookqa,7864320000.0,0.28
+GQA 2 groups,openbookqa,9437184000.0,0.286
+GQA 2 groups,openbookqa,11010048000.0,0.294
+GQA 2 groups,openbookqa,12582912000.0,0.306
+GQA 2 groups,openbookqa,14155776000.0,0.286
+GQA 2 groups,openbookqa,15728640000.0,0.302
+GQA 2 groups,openbookqa,17301504000.0,0.304
+GQA 2 groups,openbookqa,18874368000.0,0.29
+GQA 2 groups,openbookqa,20447232000.0,0.302
+GQA 2 groups,openbookqa,22020096000.0,0.31
+GQA 2 groups,openbookqa,23592960000.0,0.29
+GQA 2 groups,openbookqa,25165824000.0,0.294
+GQA 2 groups,openbookqa,26738688000.0,0.304
+GQA 2 groups,openbookqa,28311552000.0,0.308
+GQA 2 groups,openbookqa,29884416000.0,0.306
+GQA 2 groups,openbookqa,31457280000.0,0.324
+GQA 2 groups,openbookqa,33030144000.0,0.3
+GQA 2 groups,openbookqa,34603008000.0,0.306
+GQA 2 groups,openbookqa,36175872000.0,0.318
+GQA 2 groups,openbookqa,37748736000.0,0.326
+GQA 2 groups,openbookqa,39321600000.0,0.316
+GQA 2 groups,openbookqa,40894464000.0,0.316
+GQA 2 groups,openbookqa,42467328000.0,0.318
+GQA 2 groups,openbookqa,44040192000.0,0.312
+GQA 2 groups,openbookqa,45613056000.0,0.314
+GQA 2 groups,openbookqa,47185920000.0,0.332
+GQA 2 groups,openbookqa,47185920000.0,0.332
+GQA 16 groups,hellaswag,1572864000.0,0.26638119896434975
+GQA 16 groups,hellaswag,3145728000.0,0.29725154351722766
+GQA 16 groups,hellaswag,4718592000.0,0.3232423819956184
+GQA 16 groups,hellaswag,6291456000.0,0.345947022505477
+GQA 16 groups,hellaswag,7864320000.0,0.36387173869747064
+GQA 16 groups,hellaswag,9437184000.0,0.3770165305715993
+GQA 16 groups,hellaswag,11010048000.0,0.3888667596096395
+GQA 16 groups,hellaswag,12582912000.0,0.3999203345947023
+GQA 16 groups,hellaswag,14155776000.0,0.4047998406691894
+GQA 16 groups,hellaswag,15728640000.0,0.41505676160127464
+GQA 16 groups,hellaswag,17301504000.0,0.4214299940250946
+GQA 16 groups,hellaswag,18874368000.0,0.4297948615813583
+GQA 16 groups,hellaswag,20447232000.0,0.43537143995220073
+GQA 16 groups,hellaswag,22020096000.0,0.4396534554869548
+GQA 16 groups,hellaswag,23592960000.0,0.44453296156144195
+GQA 16 groups,hellaswag,25165824000.0,0.450408285202151
+GQA 16 groups,hellaswag,26738688000.0,0.4539932284405497
+GQA 16 groups,hellaswag,28311552000.0,0.4583748257319259
+GQA 16 groups,hellaswag,29884416000.0,0.4658434574785899
+GQA 16 groups,hellaswag,31457280000.0,0.4658434574785899
+GQA 16 groups,hellaswag,33030144000.0,0.46932881896036643
+GQA 16 groups,hellaswag,34603008000.0,0.4726150169288986
+GQA 16 groups,hellaswag,36175872000.0,0.4750049790878311
+GQA 16 groups,hellaswag,37748736000.0,0.47849034056960765
+GQA 16 groups,hellaswag,39321600000.0,0.47699661422027484
+GQA 16 groups,hellaswag,40894464000.0,0.4817765385381398
+GQA 16 groups,hellaswag,42467328000.0,0.483469428400717
+GQA 16 groups,hellaswag,44040192000.0,0.4838677554272057
+GQA 16 groups,hellaswag,45613056000.0,0.4871539533957379
+GQA 16 groups,hellaswag,47185920000.0,0.4876518621788488
+GQA 16 groups,hellaswag,47185920000.0,0.4876518621788488
+GQA 16 groups,mmlu,1572864000.0,0.2586058592369441
+GQA 16 groups,mmlu,3145728000.0,0.26634146971655254
+GQA 16 groups,mmlu,4718592000.0,0.2752620966818081
+GQA 16 groups,mmlu,6291456000.0,0.27723838561789776
+GQA 16 groups,mmlu,7864320000.0,0.2907593722781619
+GQA 16 groups,mmlu,9437184000.0,0.2888604138964714
+GQA 16 groups,mmlu,11010048000.0,0.2957633673200839
+GQA 16 groups,mmlu,12582912000.0,0.3028954368082722
+GQA 16 groups,mmlu,14155776000.0,0.3047139926837695
+GQA 16 groups,mmlu,15728640000.0,0.30200738396734134
+GQA 16 groups,mmlu,17301504000.0,0.2988136625121266
+GQA 16 groups,mmlu,18874368000.0,0.3049125953383158
+GQA 16 groups,mmlu,20447232000.0,0.3087076265586535
+GQA 16 groups,mmlu,22020096000.0,0.3110565537496892
+GQA 16 groups,mmlu,23592960000.0,0.31286864672604336
+GQA 16 groups,mmlu,25165824000.0,0.3165165291570455
+GQA 16 groups,mmlu,26738688000.0,0.3130227604981286
+GQA 16 groups,mmlu,28311552000.0,0.31201072875343294
+GQA 16 groups,mmlu,29884416000.0,0.31785436718480414
+GQA 16 groups,mmlu,31457280000.0,0.3174887145127641
+GQA 16 groups,mmlu,33030144000.0,0.31908027406102574
+GQA 16 groups,mmlu,34603008000.0,0.3214675989655296
+GQA 16 groups,mmlu,36175872000.0,0.31794662697281006
+GQA 16 groups,mmlu,37748736000.0,0.31853199734510235
+GQA 16 groups,mmlu,39321600000.0,0.32378989604266356
+GQA 16 groups,mmlu,40894464000.0,0.3210330675024162
+GQA 16 groups,mmlu,42467328000.0,0.3205941290618468
+GQA 16 groups,mmlu,44040192000.0,0.3229486256698073
+GQA 16 groups,mmlu,45613056000.0,0.3217503672919379
+GQA 16 groups,mmlu,47185920000.0,0.32234500341607886
+GQA 16 groups,mmlu,47185920000.0,0.32234500341607886
+GQA 16 groups,arc,1572864000.0,0.2904248687098516
+GQA 16 groups,arc,3145728000.0,0.34400245343078106
+GQA 16 groups,arc,4718592000.0,0.3685396341113064
+GQA 16 groups,arc,6291456000.0,0.37929494030176625
+GQA 16 groups,arc,7864320000.0,0.398980562163156
+GQA 16 groups,arc,9437184000.0,0.3989001218096781
+GQA 16 groups,arc,11010048000.0,0.41755079233748177
+GQA 16 groups,arc,12582912000.0,0.42658237666770094
+GQA 16 groups,arc,14155776000.0,0.4437714747015088
+GQA 16 groups,arc,15728640000.0,0.43525054297238597
+GQA 16 groups,arc,17301504000.0,0.43506883395961893
+GQA 16 groups,arc,18874368000.0,0.42441479642844826
+GQA 16 groups,arc,20447232000.0,0.44390147205846864
+GQA 16 groups,arc,22020096000.0,0.4528834993852059
+GQA 16 groups,arc,23592960000.0,0.4564437032440445
+GQA 16 groups,arc,25165824000.0,0.4564322117649763
+GQA 16 groups,arc,26738688000.0,0.4547372186024063
+GQA 16 groups,arc,28311552000.0,0.4528432792084669
+GQA 16 groups,arc,29884416000.0,0.46314323554084647
+GQA 16 groups,arc,31457280000.0,0.465936383171878
+GQA 16 groups,arc,33030144000.0,0.46427011870697876
+GQA 16 groups,arc,34603008000.0,0.4716584215304352
+GQA 16 groups,arc,36175872000.0,0.4653050700405649
+GQA 16 groups,arc,37748736000.0,0.46275683455717587
+GQA 16 groups,arc,39321600000.0,0.4657431826800427
+GQA 16 groups,arc,40894464000.0,0.4707994334700819
+GQA 16 groups,arc,42467328000.0,0.46740944714494204
+GQA 16 groups,arc,44040192000.0,0.4710385998781903
+GQA 16 groups,arc,45613056000.0,0.4723012261408166
+GQA 16 groups,arc,47185920000.0,0.4683086553820342
+GQA 16 groups,arc,47185920000.0,0.4683086553820342
+GQA 16 groups,piqa,1572864000.0,0.5620239390642002
+GQA 16 groups,piqa,3145728000.0,0.6001088139281828
+GQA 16 groups,piqa,4718592000.0,0.6137105549510338
+GQA 16 groups,piqa,6291456000.0,0.6332970620239391
+GQA 16 groups,piqa,7864320000.0,0.6485310119695321
+GQA 16 groups,piqa,9437184000.0,0.6490750816104461
+GQA 16 groups,piqa,11010048000.0,0.6686615886833515
+GQA 16 groups,piqa,12582912000.0,0.6643090315560392
+GQA 16 groups,piqa,14155776000.0,0.6681175190424374
+GQA 16 groups,piqa,15728640000.0,0.6697497279651795
+GQA 16 groups,piqa,17301504000.0,0.6779107725788901
+GQA 16 groups,piqa,18874368000.0,0.6779107725788901
+GQA 16 groups,piqa,20447232000.0,0.6871599564744287
+GQA 16 groups,piqa,22020096000.0,0.6882480957562568
+GQA 16 groups,piqa,23592960000.0,0.6833514689880305
+GQA 16 groups,piqa,25165824000.0,0.690968443960827
+GQA 16 groups,piqa,26738688000.0,0.6833514689880305
+GQA 16 groups,piqa,28311552000.0,0.6936887921653971
+GQA 16 groups,piqa,29884416000.0,0.6964091403699674
+GQA 16 groups,piqa,31457280000.0,0.7018498367791077
+GQA 16 groups,piqa,33030144000.0,0.6996735582154516
+GQA 16 groups,piqa,34603008000.0,0.7023939064200218
+GQA 16 groups,piqa,36175872000.0,0.7018498367791077
+GQA 16 groups,piqa,37748736000.0,0.7040261153427638
+GQA 16 groups,piqa,39321600000.0,0.6996735582154516
+GQA 16 groups,piqa,40894464000.0,0.6985854189336235
+GQA 16 groups,piqa,42467328000.0,0.7013057671381937
+GQA 16 groups,piqa,44040192000.0,0.7078346028291621
+GQA 16 groups,piqa,45613056000.0,0.7132752992383025
+GQA 16 groups,piqa,47185920000.0,0.7083786724700761
+GQA 16 groups,piqa,47185920000.0,0.7083786724700761
+GQA 16 groups,commonsenseqa,1572864000.0,0.2547092547092547
+GQA 16 groups,commonsenseqa,3145728000.0,0.31695331695331697
+GQA 16 groups,commonsenseqa,4718592000.0,0.3235053235053235
+GQA 16 groups,commonsenseqa,6291456000.0,0.33906633906633904
+GQA 16 groups,commonsenseqa,7864320000.0,0.3521703521703522
+GQA 16 groups,commonsenseqa,9437184000.0,0.3464373464373464
+GQA 16 groups,commonsenseqa,11010048000.0,0.3595413595413595
+GQA 16 groups,commonsenseqa,12582912000.0,0.3783783783783784
+GQA 16 groups,commonsenseqa,14155776000.0,0.3808353808353808
+GQA 16 groups,commonsenseqa,15728640000.0,0.37428337428337427
+GQA 16 groups,commonsenseqa,17301504000.0,0.3677313677313677
+GQA 16 groups,commonsenseqa,18874368000.0,0.3628173628173628
+GQA 16 groups,commonsenseqa,20447232000.0,0.3669123669123669
+GQA 16 groups,commonsenseqa,22020096000.0,0.371007371007371
+GQA 16 groups,commonsenseqa,23592960000.0,0.3857493857493858
+GQA 16 groups,commonsenseqa,25165824000.0,0.36363636363636365
+GQA 16 groups,commonsenseqa,26738688000.0,0.37182637182637185
+GQA 16 groups,commonsenseqa,28311552000.0,0.36936936936936937
+GQA 16 groups,commonsenseqa,29884416000.0,0.3824733824733825
+GQA 16 groups,commonsenseqa,31457280000.0,0.36855036855036855
+GQA 16 groups,commonsenseqa,33030144000.0,0.3701883701883702
+GQA 16 groups,commonsenseqa,34603008000.0,0.3914823914823915
+GQA 16 groups,commonsenseqa,36175872000.0,0.38001638001638
+GQA 16 groups,commonsenseqa,37748736000.0,0.3857493857493858
+GQA 16 groups,commonsenseqa,39321600000.0,0.3775593775593776
+GQA 16 groups,commonsenseqa,40894464000.0,0.37674037674037675
+GQA 16 groups,commonsenseqa,42467328000.0,0.3808353808353808
+GQA 16 groups,commonsenseqa,44040192000.0,0.38165438165438165
+GQA 16 groups,commonsenseqa,45613056000.0,0.3841113841113841
+GQA 16 groups,commonsenseqa,47185920000.0,0.38656838656838655
+GQA 16 groups,commonsenseqa,47185920000.0,0.38656838656838655
+GQA 16 groups,winogrande,1572864000.0,0.47829518547750594
+GQA 16 groups,winogrande,3145728000.0,0.48382004735595896
+GQA 16 groups,winogrande,4718592000.0,0.49171270718232046
+GQA 16 groups,winogrande,6291456000.0,0.4861878453038674
+GQA 16 groups,winogrande,7864320000.0,0.49171270718232046
+GQA 16 groups,winogrande,9437184000.0,0.48224151539068666
+GQA 16 groups,winogrande,11010048000.0,0.49329123914759276
+GQA 16 groups,winogrande,12582912000.0,0.4861878453038674
+GQA 16 groups,winogrande,14155776000.0,0.4940805051302289
+GQA 16 groups,winogrande,15728640000.0,0.4964483030781373
+GQA 16 groups,winogrande,17301504000.0,0.4925019731649566
+GQA 16 groups,winogrande,18874368000.0,0.4988161010260458
+GQA 16 groups,winogrande,20447232000.0,0.5067087608524072
+GQA 16 groups,winogrande,22020096000.0,0.505130228887135
+GQA 16 groups,winogrande,23592960000.0,0.49329123914759276
+GQA 16 groups,winogrande,25165824000.0,0.4909234411996843
+GQA 16 groups,winogrande,26738688000.0,0.4696132596685083
+GQA 16 groups,winogrande,28311552000.0,0.48855564325177586
+GQA 16 groups,winogrande,29884416000.0,0.49013417521704816
+GQA 16 groups,winogrande,31457280000.0,0.4980268350434096
+GQA 16 groups,winogrande,33030144000.0,0.4980268350434096
+GQA 16 groups,winogrande,34603008000.0,0.489344909234412
+GQA 16 groups,winogrande,36175872000.0,0.489344909234412
+GQA 16 groups,winogrande,37748736000.0,0.49171270718232046
+GQA 16 groups,winogrande,39321600000.0,0.48145224940805054
+GQA 16 groups,winogrande,40894464000.0,0.4861878453038674
+GQA 16 groups,winogrande,42467328000.0,0.49329123914759276
+GQA 16 groups,winogrande,44040192000.0,0.49171270718232046
+GQA 16 groups,winogrande,45613056000.0,0.4980268350434096
+GQA 16 groups,winogrande,47185920000.0,0.4964483030781373
+GQA 16 groups,winogrande,47185920000.0,0.4964483030781373
+GQA 16 groups,openbookqa,1572864000.0,0.264
+GQA 16 groups,openbookqa,3145728000.0,0.26
+GQA 16 groups,openbookqa,4718592000.0,0.258
+GQA 16 groups,openbookqa,6291456000.0,0.272
+GQA 16 groups,openbookqa,7864320000.0,0.268
+GQA 16 groups,openbookqa,9437184000.0,0.294
+GQA 16 groups,openbookqa,11010048000.0,0.298
+GQA 16 groups,openbookqa,12582912000.0,0.278
+GQA 16 groups,openbookqa,14155776000.0,0.288
+GQA 16 groups,openbookqa,15728640000.0,0.3
+GQA 16 groups,openbookqa,17301504000.0,0.298
+GQA 16 groups,openbookqa,18874368000.0,0.29
+GQA 16 groups,openbookqa,20447232000.0,0.31
+GQA 16 groups,openbookqa,22020096000.0,0.316
+GQA 16 groups,openbookqa,23592960000.0,0.312
+GQA 16 groups,openbookqa,25165824000.0,0.306
+GQA 16 groups,openbookqa,26738688000.0,0.31
+GQA 16 groups,openbookqa,28311552000.0,0.314
+GQA 16 groups,openbookqa,29884416000.0,0.322
+GQA 16 groups,openbookqa,31457280000.0,0.312
+GQA 16 groups,openbookqa,33030144000.0,0.308
+GQA 16 groups,openbookqa,34603008000.0,0.314
+GQA 16 groups,openbookqa,36175872000.0,0.312
+GQA 16 groups,openbookqa,37748736000.0,0.312
+GQA 16 groups,openbookqa,39321600000.0,0.318
+GQA 16 groups,openbookqa,40894464000.0,0.31
+GQA 16 groups,openbookqa,42467328000.0,0.308
+GQA 16 groups,openbookqa,44040192000.0,0.31
+GQA 16 groups,openbookqa,45613056000.0,0.318
+GQA 16 groups,openbookqa,47185920000.0,0.314
+GQA 16 groups,openbookqa,47185920000.0,0.314

data/banner_visualisation_data.csv

@@ -0,0 +1,1575 @@
+original_id,category,subset,x_position,y_position
+0,Chart & Table,CoSyn_400k_chart,-5.8737874,15.983195
+1,Chart & Table,CoSyn_400k_chart,-4.945642,16.106464
+2,Chart & Table,CoSyn_400k_chart,-5.5937595,16.452412
+3,Science,CoSyn_400k_chemical,-1.5379021,10.981212
+4,OCR QA,CoSyn_400k_diagram,-5.065811,13.181871
+5,General VQA,CoSyn_400k_graphic,-0.48748758,15.743629
+6,Chart & Table,Unichart,-5.478337,15.044097
+7,General VQA,LLaVA_Instruct_150K,-0.04121901,13.326656
+original_id,category,subset,x_position,y_position
+0,Chart & Table,CoSyn_400k_chart,-5.8737874,15.983195
+1,Chart & Table,CoSyn_400k_chart,-4.945642,16.106464
+2,Chart & Table,CoSyn_400k_chart,-5.5937595,16.452412
+3,Chart & Table,CoSyn_400k_chart,-6.8471727,15.6502905
+4,Chart & Table,CoSyn_400k_chart,-4.593728,15.400246
+5,Chart & Table,CoSyn_400k_chart,-6.5247335,15.286765
+6,Chart & Table,CoSyn_400k_chart,-5.9858575,13.354458
+7,Chart & Table,CoSyn_400k_chart,-5.9055824,15.795122
+8,Chart & Table,CoSyn_400k_chart,-4.824333,15.338556
+9,Chart & Table,CoSyn_400k_chart,-6.677866,14.968742
+10,Chart & Table,CoSyn_400k_chart,-5.6846857,16.26988
+11,Science,CoSyn_400k_chemical,-1.5379021,10.981212
+12,Science,CoSyn_400k_circuit,-5.0486145,14.184427
+13,OCR QA,CoSyn_400k_diagram,-5.065811,13.181871
+14,OCR QA,CoSyn_400k_diagram,-5.261097,14.098007
+15,OCR QA,CoSyn_400k_diagram,-4.6068473,14.401655
+16,OCR QA,CoSyn_400k_document,-3.903585,13.690944
+17,OCR QA,CoSyn_400k_document,-3.875524,17.058977
+18,OCR QA,CoSyn_400k_document,-3.7943916,17.480606
+19,OCR QA,CoSyn_400k_document,-4.4624634,17.283459
+20,OCR QA,CoSyn_400k_document,-1.7727383,18.533064
+21,OCR QA,CoSyn_400k_document,-2.8321483,17.436039
+22,OCR QA,CoSyn_400k_document,-6.461951,14.257359
+23,General VQA,CoSyn_400k_graphic,-5.8664246,16.971214
+24,General VQA,CoSyn_400k_graphic,-0.48748758,15.743629
+25,Mathematics,CoSyn_400k_math,-0.03838683,16.25767
+26,Mathematics,CoSyn_400k_math,-0.4915684,16.225067
+27,Mathematics,CoSyn_400k_math,-1.1946124,16.545416
+28,Mathematics,CoSyn_400k_math,-0.534912,16.165432
+29,Mathematics,CoSyn_400k_math,0.022691553,16.158678
+30,Mathematics,CoSyn_400k_math,0.40876907,16.072973
+31,OCR QA,CoSyn_400k_music,-4.5062866,14.961522
+32,OCR QA,CoSyn_400k_nutrition,-4.331227,16.355286
+33,Chart & Table,CoSyn_400k_table,-5.3820634,16.286442
+34,Chart & Table,CoSyn_400k_table,-4.434908,16.792274
+35,Chart & Table,CoSyn_400k_table,-6.117985,15.993164
+36,Chart & Table,CoSyn_400k_table,-4.9423914,16.244167
+37,OCR QA,DoclingMatix,-2.123511,19.758589
+38,OCR QA,DoclingMatix,-1.2339255,18.348934
+39,OCR QA,DoclingMatix,-2.2721937,17.460812
+40,OCR QA,DoclingMatix,-1.4948343,19.09816
+41,OCR QA,DoclingMatix,-2.001817,18.948124
+42,OCR QA,DoclingMatix,-3.313269,18.66045
+43,OCR QA,DoclingMatix,-3.0790706,16.690575
+44,OCR QA,DoclingMatix,-2.2622058,18.828022
+45,OCR QA,DoclingMatix,-2.583864,18.676437
+46,OCR QA,DoclingMatix,-6.403138,14.5712185
+47,OCR QA,DoclingMatix,-2.0187643,19.280146
+48,OCR QA,DoclingMatix,-1.7590756,18.563002
+49,OCR QA,DoclingMatix,-1.2618425,17.422573
+50,OCR QA,DoclingMatix,-3.6012855,18.275747
+51,OCR QA,DoclingMatix,-2.6068192,17.872808
+52,OCR QA,DoclingMatix,-3.283061,18.019083
+53,OCR QA,DoclingMatix,-3.0331342,18.203625
+54,OCR QA,DoclingMatix,-2.2425787,19.199886
+55,OCR QA,DoclingMatix,-1.941939,19.540573
+56,OCR QA,DoclingMatix,-2.0318031,15.930559
+57,OCR QA,DoclingMatix,-2.3450463,18.587273
+58,OCR QA,DoclingMatix,-1.4967664,18.750023
+59,OCR QA,DoclingMatix,-3.4411478,17.345871
+60,OCR QA,DoclingMatix,-1.6649085,19.57321
+61,OCR QA,DoclingMatix,-2.381297,19.393425
+62,OCR QA,DoclingMatix,-2.3868053,18.25288
+63,OCR QA,DoclingMatix,-3.477942,18.536766
+64,OCR QA,DoclingMatix,-3.3863645,18.042032
+65,OCR QA,DoclingMatix,-3.381429,19.029367
+66,OCR QA,DoclingMatix,-2.0542133,19.27429
+67,OCR QA,DoclingMatix,-2.0553746,17.631887
+68,OCR QA,DoclingMatix,-1.8904015,18.499573
+69,OCR QA,DoclingMatix,-2.3417099,18.13063
+70,OCR QA,DoclingMatix,-2.1454659,18.419767
+71,OCR QA,DoclingMatix,-2.7915456,19.205055
+72,OCR QA,DoclingMatix,-3.7376049,16.510166
+73,OCR QA,DoclingMatix,-2.171803,17.909151
+74,OCR QA,DoclingMatix,-2.4062195,19.129288
+75,OCR QA,DoclingMatix,-3.1809347,18.645334
+76,OCR QA,DoclingMatix,-2.0148678,19.54656
+77,OCR QA,DoclingMatix,-2.2982774,18.211866
+78,OCR QA,DoclingMatix,-1.4421009,16.974617
+79,OCR QA,DoclingMatix,-1.792398,17.167984
+80,OCR QA,DoclingMatix,-2.3227503,18.35669
+81,OCR QA,DoclingMatix,-1.9308434,15.982081
+82,OCR QA,DoclingMatix,-3.4728649,18.326372
+83,OCR QA,DoclingMatix,-1.9811486,19.728716
+84,OCR QA,DoclingMatix,-2.9568808,18.825401
+85,OCR QA,DoclingMatix,-4.254185,13.623234
+86,OCR QA,DoclingMatix,-4.4296207,15.911434
+87,OCR QA,DoclingMatix,-2.482577,17.241879
+88,OCR QA,DoclingMatix,-1.9588791,19.403658
+89,OCR QA,DoclingMatix,-1.292695,17.806475
+90,OCR QA,DoclingMatix,-2.3372555,18.205208
+91,OCR QA,DoclingMatix,-2.8207717,19.285505
+92,OCR QA,DoclingMatix,-1.9746872,19.76563
+93,OCR QA,DoclingMatix,-1.8576025,19.721828
+94,OCR QA,DoclingMatix,-1.5742756,19.052324
+95,OCR QA,DoclingMatix,-1.8536038,19.753435
+96,OCR QA,DoclingMatix,-1.763717,17.355995
+97,OCR QA,DoclingMatix,-2.303972,17.225918
+98,OCR QA,DoclingMatix,-1.8316678,18.82974
+99,OCR QA,DoclingMatix,-1.5295701,19.64252
+100,OCR QA,DoclingMatix,-0.9577445,17.73931
+101,OCR QA,DoclingMatix,-3.5142593,18.210361
+102,OCR QA,DoclingMatix,-1.527282,17.49759
+103,OCR QA,DoclingMatix,-2.3261003,18.39049
+104,OCR QA,DoclingMatix,-2.4651723,19.005152
+105,OCR QA,DoclingMatix,-4.523276,15.443817
+106,OCR QA,DoclingMatix,-1.5367091,18.813587
+107,OCR QA,DoclingMatix,-2.3303423,18.057833
+108,OCR QA,DoclingMatix,-2.1394675,18.376097
+109,OCR QA,DoclingMatix,-2.3630292,19.395987
+110,OCR QA,DoclingMatix,-1.6529044,19.392685
+111,OCR QA,DoclingMatix,-2.4107027,16.50055
+112,OCR QA,DoclingMatix,-5.4095883,16.071615
+113,OCR QA,DoclingMatix,-3.3529801,18.519686
+114,OCR QA,DoclingMatix,-2.9411347,18.826674
+115,OCR QA,DoclingMatix,-2.1890736,18.415182
+116,OCR QA,DoclingMatix,-2.2139513,18.754198
+117,OCR QA,DoclingMatix,-3.2719424,19.08318
+118,OCR QA,DoclingMatix,-2.8656187,16.53646
+119,OCR QA,DoclingMatix,-1.6826456,18.750132
+120,OCR QA,DoclingMatix,-1.482793,18.30256
+121,OCR QA,DoclingMatix,-1.4262551,19.035845
+122,OCR QA,DoclingMatix,-2.6542027,19.080572
+123,OCR QA,DoclingMatix,-1.733341,19.370384
+124,OCR QA,DoclingMatix,-2.5386746,18.128016
+125,OCR QA,DoclingMatix,-1.9642893,18.426172
+126,OCR QA,DoclingMatix,-2.4258876,19.42167
+127,OCR QA,DoclingMatix,-4.8599396,15.642522
+128,OCR QA,DoclingMatix,-1.4616601,18.574923
+129,OCR QA,DoclingMatix,-2.0061698,18.075985
+130,OCR QA,DoclingMatix,-2.2746208,19.78703
+131,OCR QA,DoclingMatix,-2.6560168,17.990816
+132,OCR QA,DoclingMatix,-3.1165454,18.33594
+133,OCR QA,DoclingMatix,-1.6757885,17.211813
+134,OCR QA,DoclingMatix,-1.4582803,18.53178
+135,OCR QA,DoclingMatix,-2.4365861,16.4549
+136,OCR QA,DoclingMatix,-1.9898921,18.799305
+137,OCR QA,DoclingMatix,-2.3840768,19.101894
+138,OCR QA,DoclingMatix,-2.6806266,17.765152
+139,OCR QA,DoclingMatix,-1.2126298,18.865656
+140,OCR QA,DoclingMatix,-3.1112385,18.96933
+141,OCR QA,DoclingMatix,-1.4632803,17.64587
+142,OCR QA,DoclingMatix,-1.893244,18.17506
+143,OCR QA,DoclingMatix,-1.6842548,17.662289
+144,OCR QA,DoclingMatix,-3.3209038,17.676641
+145,OCR QA,DoclingMatix,-3.5430915,18.633417
+146,OCR QA,DoclingMatix,-1.4177037,17.525536
+147,OCR QA,DoclingMatix,-1.7031838,18.482338
+148,OCR QA,DoclingMatix,-2.6439989,18.788374
+149,OCR QA,DoclingMatix,-2.597876,19.185583
+150,OCR QA,DoclingMatix,-0.5687146,17.854235
+151,OCR QA,DoclingMatix,-1.5822437,17.89997
+152,OCR QA,DoclingMatix,-2.5681157,15.862703
+153,OCR QA,DoclingMatix,-3.1931202,18.610592
+154,OCR QA,DoclingMatix,-2.2833252,16.032545
+155,OCR QA,DoclingMatix,-2.7539678,18.959908
+156,OCR QA,DoclingMatix,-3.293064,18.738478
+157,OCR QA,DoclingMatix,-4.458434,13.078367
+158,OCR QA,DoclingMatix,-2.1089303,16.492489
+159,OCR QA,DoclingMatix,-2.9728808,18.275713
+160,OCR QA,DoclingMatix,-1.9504831,17.425997
+161,OCR QA,DoclingMatix,-2.1972477,18.952415
+162,OCR QA,DoclingMatix,-1.9784061,18.167784
+163,OCR QA,DoclingMatix,-2.1374834,18.780544
+164,General VQA,LLaVA_Instruct_150K,-0.04121901,13.326656
+165,General VQA,LLaVA_Instruct_150K,-3.003669,13.9813
+166,General VQA,LLaVA_Instruct_150K,-0.87659425,15.0002365
+167,General VQA,LLaVA_Instruct_150K,-2.4315653,13.461026
+168,General VQA,LLaVA_Instruct_150K,-3.657344,13.454892
+169,General VQA,LLaVA_Instruct_150K,-3.3701751,13.903656
+170,General VQA,LLaVA_Instruct_150K,-1.4356084,12.906051
+171,General VQA,LLaVA_Instruct_150K,-2.0306752,13.837592
+172,General VQA,LLaVA_Instruct_150K,-1.93058,13.64789
+173,General VQA,LLaVA_Instruct_150K,-2.960764,12.389478
+174,General VQA,LLaVA_Instruct_150K,-1.5670083,12.371494
+175,General VQA,LLaVA_Instruct_150K,-1.3098346,15.785713
+176,General VQA,LLaVA_Instruct_150K,-4.1270065,15.013622
+177,General VQA,LLaVA_Instruct_150K,-2.6499448,13.135092
+178,General VQA,LLaVA_Instruct_150K,-1.9295284,16.038881
+179,Chart & Table,SynthChartNet,-5.1461444,17.769293
+180,Chart & Table,SynthChartNet,-6.727832,15.511663
+181,Chart & Table,SynthChartNet,-6.411554,16.44527
+182,Chart & Table,SynthChartNet,-7.151805,14.436375
+183,Chart & Table,SynthChartNet,-6.6720834,16.477493
+184,Chart & Table,SynthChartNet,-6.1793976,15.210242
+185,Chart & Table,SynthChartNet,-5.141068,17.590567
+186,Chart & Table,SynthChartNet,-7.0122495,15.615927
+187,Chart & Table,SynthChartNet,-6.944416,15.583528
+188,Chart & Table,SynthChartNet,-5.294619,18.07463
+189,Chart & Table,SynthChartNet,-5.207549,17.901262
+190,Chart & Table,SynthChartNet,-5.176415,18.117426
+191,Chart & Table,SynthChartNet,-5.208407,18.033777
+192,Chart & Table,SynthChartNet,-5.368511,17.664892
+193,Chart & Table,SynthChartNet,-5.2063956,17.942322
+194,Chart & Table,SynthChartNet,-6.985873,16.031464
+195,Chart & Table,SynthChartNet,-6.5159984,15.32247
+196,Chart & Table,SynthChartNet,-6.9621744,16.361357
+197,Chart & Table,SynthChartNet,-5.2207456,17.69036
+198,Chart & Table,SynthChartNet,-6.8150105,16.328646
+199,Chart & Table,SynthChartNet,-5.163355,18.21635
+200,Chart & Table,SynthChartNet,-6.920904,16.498514
+201,Chart & Table,SynthChartNet,-6.4768267,16.19447
+202,Chart & Table,SynthChartNet,-6.704763,16.307007
+203,Chart & Table,SynthChartNet,-6.8234115,16.408567
+204,Chart & Table,SynthChartNet,-5.481283,17.856428
+205,Chart & Table,SynthChartNet,-6.543561,16.619896
+206,Chart & Table,SynthChartNet,-4.852024,18.189837
+207,Chart & Table,SynthChartNet,-5.215617,17.742992
+208,Chart & Table,SynthChartNet,-6.4959555,16.68928
+209,Chart & Table,SynthChartNet,-5.388788,17.987978
+210,Chart & Table,SynthChartNet,-6.4665,15.409831
+211,Chart & Table,SynthChartNet,-6.5481486,16.35738
+212,Chart & Table,SynthChartNet,-5.4961696,17.997698
+213,Chart & Table,SynthChartNet,-5.5057435,17.749802
+214,Chart & Table,SynthChartNet,-5.2866464,17.925413
+215,Chart & Table,SynthChartNet,-6.575139,15.787562
+216,Chart & Table,SynthChartNet,-5.6337447,17.884447
+217,Chart & Table,SynthChartNet,-5.428942,18.140358
+218,Chart & Table,SynthChartNet,-5.6336236,17.739908
+219,Chart & Table,SynthChartNet,-6.3776565,16.402546
+220,Chart & Table,SynthChartNet,-5.564274,17.744778
+221,Chart & Table,SynthChartNet,-6.9941792,16.175684
+222,Chart & Table,SynthChartNet,-6.6928587,16.116158
+223,Chart & Table,SynthChartNet,-6.7574,15.715918
+224,Chart & Table,SynthChartNet,-6.9718375,16.300568
+225,Chart & Table,SynthChartNet,-6.9625235,16.005453
+226,Chart & Table,SynthChartNet,-7.0678654,15.965807
+227,Chart & Table,SynthChartNet,-5.0787272,18.141487
+228,Chart & Table,SynthChartNet,-5.0736647,18.061468
+229,Naive OCR,SynthCodeNet,0.04656302,17.572514
+230,Naive OCR,SynthCodeNet,-1.1099467,17.690296
+231,Naive OCR,SynthCodeNet,-0.04873978,17.705694
+232,Naive OCR,SynthCodeNet,-0.33281502,16.677038
+233,Naive OCR,SynthCodeNet,1.0885636,15.638341
+234,Naive OCR,SynthCodeNet,-4.2552023,15.99757
+235,Naive OCR,SynthCodeNet,-3.852895,15.5951185
+236,Naive OCR,SynthCodeNet,-0.3341921,17.734613
+237,Naive OCR,SynthCodeNet,0.42134494,16.38415
+238,Naive OCR,SynthCodeNet,-0.42474476,17.735378
+239,Naive OCR,SynthCodeNet,-0.26788715,17.153187
+240,Naive OCR,SynthCodeNet,-0.19620547,18.005255
+241,Naive OCR,SynthCodeNet,-0.058252852,17.238487
+242,Naive OCR,SynthCodeNet,-1.7247115,18.295624
+243,Naive OCR,SynthCodeNet,-0.009917033,17.808311
+244,Naive OCR,SynthCodeNet,-1.387855,17.145159
+245,Naive OCR,SynthCodeNet,-1.701879,17.920927
+246,Naive OCR,SynthCodeNet,-3.6778998,16.088854
+247,Naive OCR,SynthCodeNet,-0.24548063,17.196598
+248,Naive OCR,SynthCodeNet,0.11758641,17.863241
+249,Naive OCR,SynthCodeNet,-1.2178589,18.26013
+250,Naive OCR,SynthCodeNet,0.0003795521,17.943213
+251,Naive OCR,SynthCodeNet,-0.42465442,18.043238
+252,Naive OCR,SynthCodeNet,-0.07526772,18.020924
+253,Naive OCR,SynthCodeNet,-0.14518633,17.87866
+254,Naive OCR,SynthCodeNet,-0.49168557,16.688969
+255,Naive OCR,SynthCodeNet,-0.31747556,18.093529
+256,Naive OCR,SynthCodeNet,0.80161583,15.814559
+257,Naive OCR,SynthCodeNet,-4.310662,16.584866
+258,Naive OCR,SynthCodeNet,-0.00977412,16.915396
+259,Naive OCR,SynthCodeNet,-4.188083,16.494669
+260,Naive OCR,SynthCodeNet,-0.25779262,18.010223
+261,Naive OCR,SynthCodeNet,-3.9593086,15.766136
+262,Naive OCR,SynthCodeNet,-1.3795527,18.213293
+263,Naive OCR,SynthCodeNet,-0.45508116,17.52061
+264,Naive OCR,SynthCodeNet,-0.2640966,17.886843
+265,Naive OCR,SynthCodeNet,-6.582913,14.3813305
+266,Naive OCR,SynthCodeNet,-4.188884,16.041538
+267,Naive OCR,SynthCodeNet,-0.32475826,17.145645
+268,Naive OCR,SynthCodeNet,-3.9861093,15.823854
+269,Naive OCR,SynthCodeNet,-0.9563473,18.1237
+270,Naive OCR,SynthCodeNet,-1.520535,18.000816
+271,Naive OCR,SynthCodeNet,-0.42804614,17.700657
+272,Naive OCR,SynthCodeNet,-1.2626554,18.494741
+273,Naive OCR,SynthCodeNet,-0.11364583,17.126024
+274,Naive OCR,SynthCodeNet,-0.34699938,17.888674
+275,Naive OCR,SynthCodeNet,-0.015610901,17.330435
+276,Naive OCR,SynthCodeNet,-3.8931282,15.892088
+277,Naive OCR,SynthCodeNet,-0.08240982,17.493
+278,Naive OCR,SynthFormulaNet,0.22886933,15.002758
+279,Naive OCR,SynthFormulaNet,1.7862303,14.29242
+280,Naive OCR,SynthFormulaNet,1.4039996,14.474301
+281,Naive OCR,SynthFormulaNet,1.7801038,14.159131
+282,Naive OCR,SynthFormulaNet,2.364047,14.913048
+283,Naive OCR,SynthFormulaNet,2.067444,13.650322
+284,Naive OCR,SynthFormulaNet,-3.9578013,12.279434
+285,Naive OCR,SynthFormulaNet,1.2084179,14.285225
+286,Naive OCR,SynthFormulaNet,1.4971747,13.390962
+287,Naive OCR,SynthFormulaNet,1.684946,14.628642
+288,Naive OCR,SynthFormulaNet,1.3491838,13.396536
+289,Naive OCR,SynthFormulaNet,1.8627237,13.770465
+290,Naive OCR,SynthFormulaNet,1.0238231,14.866128
+291,Naive OCR,SynthFormulaNet,1.364893,14.193316
+292,Naive OCR,SynthFormulaNet,1.4466103,13.644283
+293,Naive OCR,SynthFormulaNet,1.0992525,14.317199
+294,Naive OCR,SynthFormulaNet,1.5460949,15.546643
+295,Naive OCR,SynthFormulaNet,1.2137767,13.969349
+296,Naive OCR,SynthFormulaNet,1.4102746,13.717272
+297,Naive OCR,SynthFormulaNet,0.91062367,13.0298605
+298,Naive OCR,SynthFormulaNet,1.1339371,13.871966
+299,Naive OCR,SynthFormulaNet,1.6997597,15.142788
+300,Naive OCR,SynthFormulaNet,0.96059656,12.709436
+301,Naive OCR,SynthFormulaNet,1.6072994,13.246107
+302,Naive OCR,SynthFormulaNet,1.1215231,13.370174
+303,Naive OCR,SynthFormulaNet,1.492844,13.373532
+304,Naive OCR,SynthFormulaNet,1.7269921,13.755263
+305,Naive OCR,SynthFormulaNet,2.179824,14.5509815
+306,Naive OCR,SynthFormulaNet,1.3578401,14.855309
+307,Naive OCR,SynthFormulaNet,1.2492148,14.366605
+308,Naive OCR,SynthFormulaNet,1.9348897,13.490846
+309,Naive OCR,SynthFormulaNet,1.407853,14.599283
+310,Naive OCR,SynthFormulaNet,1.3527994,14.69017
+311,Naive OCR,SynthFormulaNet,1.0525184,12.86483
+312,Naive OCR,SynthFormulaNet,1.9524969,13.247496
+313,Naive OCR,SynthFormulaNet,1.1278163,14.295946
+314,Naive OCR,SynthFormulaNet,2.24632,13.667028
+315,Naive OCR,SynthFormulaNet,1.5239518,14.299738
+316,Naive OCR,SynthFormulaNet,1.7567451,13.165833
+317,Naive OCR,SynthFormulaNet,1.140315,13.921025
+318,Naive OCR,SynthFormulaNet,1.6855685,15.429556
+319,Naive OCR,SynthFormulaNet,1.1004134,13.143863
+320,Naive OCR,SynthFormulaNet,1.4039396,14.001242
+321,Naive OCR,SynthFormulaNet,1.4621967,14.645452
+322,Naive OCR,SynthFormulaNet,1.2914104,14.500753
+323,Naive OCR,SynthFormulaNet,1.8754648,14.831985
+324,Naive OCR,SynthFormulaNet,1.6577595,15.188357
+325,Naive OCR,SynthFormulaNet,1.6267941,13.784543
+326,Naive OCR,SynthFormulaNet,1.6241556,13.506343
+327,Chart & Table,Unichart,-5.478337,15.044097
+328,Chart & Table,Unichart,-5.5799456,13.085751
+329,Chart & Table,Unichart,-5.949499,14.996578
+330,Chart & Table,Unichart,-5.516415,13.015133
+331,Chart & Table,Unichart,-6.098518,13.201147
+332,Chart & Table,Unichart,-6.9667177,14.026482
+333,Chart & Table,Unichart,-6.322132,13.28759
+334,Chart & Table,Unichart,-6.565431,15.262299
+335,Chart & Table,Unichart,-7.018406,14.05617
+336,Chart & Table,Unichart,-6.7399693,13.898848
+337,Chart & Table,Unichart,-5.7210565,13.72224
+338,Chart & Table,Unichart,-5.632763,14.958898
+339,Chart & Table,Unichart,-6.947203,14.937348
+340,Chart & Table,Unichart,-5.8241777,13.499949
+341,Chart & Table,Unichart,-6.3517947,16.035973
+342,Chart & Table,Unichart,-5.290241,17.525455
+343,Chart & Table,Unichart,-5.9058266,14.941877
+344,Chart & Table,Unichart,-4.604956,17.870945
+345,Chart & Table,Unichart,-5.79115,12.908007
+346,Chart & Table,Unichart,-6.2544575,14.887673
+347,Chart & Table,Unichart,-5.753358,15.010925
+348,Chart & Table,Unichart,-5.786733,13.154917
+349,Chart & Table,Unichart,-5.8034973,12.8242655
+350,Chart & Table,Unichart,-6.31294,16.048944
+351,Chart & Table,Unichart,-7.2014155,15.414163
+352,Chart & Table,Unichart,-6.2808485,16.195747
+353,Chart & Table,Unichart,-5.985574,15.921576
+354,Chart & Table,Unichart,-7.360994,15.127242
+355,Chart & Table,Unichart,-5.8023033,14.888992
+356,Chart & Table,Unichart,-5.602459,14.954068
+357,Chart & Table,Unichart,-6.262162,14.695169
+358,Chart & Table,Unichart,-7.0918374,14.893238
+359,Chart & Table,Unichart,-6.868848,14.455841
+360,Chart & Table,Unichart,-5.7695827,15.218684
+361,Chart & Table,Unichart,-5.7917295,14.848716
+362,Chart & Table,Unichart,-5.606748,15.595446
+363,Chart & Table,Unichart,-7.3670974,14.436186
+364,Chart & Table,Unichart,-6.102013,13.038253
+365,Chart & Table,Unichart,-5.6951466,14.418335
+366,Chart & Table,Unichart,-6.119423,13.288748
+367,Chart & Table,Unichart,-6.938326,13.939838
+368,Chart & Table,Unichart,-6.1431007,16.141819
+369,Chart & Table,Unichart,-5.6071377,15.208249
+370,Chart & Table,Unichart,-5.768991,15.128817
+371,Chart & Table,Unichart,-5.259301,17.546667
+372,Chart & Table,Unichart,-4.77272,17.328728
+373,Chart & Table,Unichart,-5.8661036,13.971627
+374,Chart & Table,Unichart,-6.908078,14.750717
+375,Chart & Table,Unichart,-6.353714,15.989009
+376,Chart & Table,Unichart,-5.8717313,14.890253
+377,Chart & Table,Unichart,-7.229736,15.591386
+378,Chart & Table,Unichart,-6.3437076,13.1844015
+379,Chart & Table,Unichart,-5.798029,14.378434
+380,Chart & Table,Unichart,-5.3162193,14.234495
+381,Chart & Table,Unichart,-6.373933,13.344297
+382,Chart & Table,Unichart,-7.078772,14.764631
+383,Chart & Table,Unichart,-5.9133935,12.824556
+384,Chart & Table,Unichart,-6.828757,14.082631
+385,Chart & Table,Unichart,-6.7189755,14.999983
+386,Chart & Table,Unichart,-6.0642824,16.449152
+387,Chart & Table,Unichart,-6.6790447,13.80666
+388,OCR QA,a_okvqa,-0.5733542,12.801027
+389,OCR QA,a_okvqa,-1.7158192,10.843025
+390,OCR QA,a_okvqa,-3.2833407,12.822911
+391,OCR QA,a_okvqa,-0.80873895,14.536217
+392,OCR QA,a_okvqa,-3.208334,14.578856
+393,Grounding & Counting,aguvis-stage-1,-3.92464,17.013128
+394,Grounding & Counting,aguvis-stage-1,-3.1343591,16.968803
+395,Grounding & Counting,aguvis-stage-1,-3.9606252,16.346624
+396,Grounding & Counting,aguvis-stage-1,-3.0290737,16.770638
+397,Grounding & Counting,aguvis-stage-1,-1.1558208,17.359201
+398,Grounding & Counting,aguvis-stage-1,-6.418879,16.79059
+399,Grounding & Counting,aguvis-stage-1,-2.3867943,15.352943
+400,Grounding & Counting,aguvis-stage-1,-2.6866517,16.432796
+401,Grounding & Counting,aguvis-stage-1,-2.380073,17.779228
+402,Grounding & Counting,aguvis-stage-1,-3.8589,17.059746
+403,Grounding & Counting,aguvis-stage-1,-3.3082666,17.71559
+404,Grounding & Counting,aguvis-stage-1,-2.5477545,16.541817
+405,Grounding & Counting,aguvis-stage-1,-2.3461995,16.683966
+406,Grounding & Counting,aguvis-stage-1,-2.909816,17.15795
+407,Grounding & Counting,aguvis-stage-1,-3.8193007,16.196898
+408,Grounding & Counting,aguvis-stage-1,-1.8479805,18.61108
+409,Grounding & Counting,aguvis-stage-1,-3.0331044,17.226364
+410,Grounding & Counting,aguvis-stage-1,-2.9594963,15.341776
+411,Grounding & Counting,aguvis-stage-1,-2.7642896,17.29218
+412,Grounding & Counting,aguvis-stage-1,-4.397569,16.361464
+413,Grounding & Counting,aguvis-stage-1,-2.588429,15.981421
+414,Grounding & Counting,aguvis-stage-1,-2.4921746,16.337145
+415,Grounding & Counting,aguvis-stage-1,-1.634949,18.328835
+416,Grounding & Counting,aguvis-stage-1,-3.5193944,17.553333
+417,Grounding & Counting,aguvis-stage-1,1.3164409,12.131497
+418,Grounding & Counting,aguvis-stage-1,-3.9198072,16.698154
+419,Grounding & Counting,aguvis-stage-1,-2.6971705,17.27412
+420,Grounding & Counting,aguvis-stage-1,-2.6077929,16.321354
+421,Grounding & Counting,aguvis-stage-1,-3.6373217,16.949177
+422,Grounding & Counting,aguvis-stage-1,-3.1801188,17.075556
+423,Grounding & Counting,aguvis-stage-1,-2.3237062,17.239109
+424,Grounding & Counting,aguvis-stage-1,-3.7926881,16.6358
+425,Grounding & Counting,aguvis-stage-1,-2.97919,15.463035
+426,Grounding & Counting,aguvis-stage-1,-3.2354767,16.985542
+427,Grounding & Counting,aguvis-stage-1,-2.9599688,17.437643
+428,Grounding & Counting,aguvis-stage-1,-2.1633608,17.812645
+429,Grounding & Counting,aguvis-stage-1,-2.6721504,16.597218
+430,Grounding & Counting,aguvis-stage-1,-3.8145788,17.693523
+431,Grounding & Counting,aguvis-stage-1,-3.021984,15.80835
+432,Grounding & Counting,aguvis-stage-1,-2.8769467,16.994005
+433,Grounding & Counting,aguvis-stage-1,-3.156238,17.429504
+434,Grounding & Counting,aguvis-stage-1,-4.332312,17.716543
+435,Grounding & Counting,aguvis-stage-1,-3.4098995,16.634125
+436,Grounding & Counting,aguvis-stage-1,-3.658236,18.488964
+437,Grounding & Counting,aguvis-stage-1,-3.019246,18.18975
+438,Science,ai2d_merged,-5.1963496,16.62253
+439,General VQA,alfworldgpt,-3.1402702,15.006798
+440,General VQA,alfworldgpt,-3.655326,14.881633
+441,General VQA,alfworldgpt,-2.9178991,13.226537
+442,General VQA,alfworldgpt,-3.446364,14.249417
+443,General VQA,allava_laion,-2.8572686,13.270778
+444,General VQA,allava_laion,-1.558349,13.590846
+445,General VQA,allava_laion,-4.4869156,13.266971
+446,General VQA,allava_laion,-0.56136036,14.288783
+447,General VQA,allava_laion,-1.9938369,16.406837
+448,General VQA,allava_laion,-2.9673812,14.970761
+449,General VQA,allava_laion,-2.055943,12.913793
+450,General VQA,allava_laion,-2.1113553,16.12969
+451,General VQA,allava_laion,-4.0677795,14.782865
+452,General VQA,allava_laion,-5.437868,14.414292
+453,General VQA,allava_laion,-1.4492984,16.317406
+454,General VQA,allava_laion,-3.0221918,16.395374
+455,General VQA,allava_laion,-1.9317902,13.783306
+456,General VQA,allava_laion,-2.6455994,16.673138
+457,General VQA,allava_laion,-2.9869754,15.000495
+458,General VQA,allava_laion,-2.3168645,14.673664
+459,General VQA,allava_laion,-1.2214115,13.521346
+460,General VQA,allava_laion,-4.027829,14.150317
+461,General VQA,allava_laion,-4.99002,17.623316
+462,General VQA,allava_laion,-1.7191764,13.515034
+463,General VQA,allava_laion,-2.7818463,15.766099
+464,General VQA,allava_laion,-1.2120026,13.573033
+465,General VQA,allava_laion,-2.9067914,14.31069
+466,General VQA,allava_laion,-1.7481589,15.501534
+467,General VQA,allava_laion,-1.4285904,13.622949
+468,General VQA,allava_laion,-2.508318,14.214216
+469,General VQA,allava_laion,-2.8544402,15.464485
+470,General VQA,allava_laion,-1.517965,12.944219
+471,General VQA,allava_laion,0.66665745,13.841945
+472,General VQA,allava_laion,-3.3896854,15.587829
+473,General VQA,allava_laion,-2.0731955,14.096371
+474,General VQA,allava_laion,-2.163503,13.447432
+475,General VQA,allava_laion,-1.7301064,14.63413
+476,General VQA,allava_laion,-2.063952,13.306315
+477,General VQA,allava_laion,-1.9544251,13.65066
+478,General VQA,allava_laion,-4.666468,14.2345295
+479,General VQA,allava_laion,-3.433153,15.611582
+480,General VQA,allava_laion,-1.3665693,13.968209
+481,General VQA,allava_laion,-2.0089586,12.901301
+482,General VQA,allava_laion,-2.5629804,13.241218
+483,General VQA,allava_laion,-1.879719,14.757591
+484,General VQA,allava_laion,-1.976303,15.084605
+485,General VQA,allava_laion,-2.668102,16.139841
+486,General VQA,allava_laion,-2.8400748,13.738467
+487,General VQA,allava_laion,-1.5293812,16.528149
+488,General VQA,allava_laion,-4.2499733,15.143023
+489,General VQA,allava_vflan,-1.3735987,13.102209
+490,General VQA,allava_vflan,-1.9468843,16.564497
+491,General VQA,allava_vflan,-4.597105,13.285384
+492,General VQA,allava_vflan,-2.4476461,17.51059
+493,General VQA,allava_vflan,-1.7535806,12.742285
+494,General VQA,allava_vflan,-2.0745947,10.439247
+495,General VQA,allava_vflan,0.7946193,12.369304
+496,General VQA,allava_vflan,-3.6089487,14.947047
+497,General VQA,allava_vflan,-2.1644802,15.040754
+498,General VQA,allava_vflan,-0.87625706,15.047502
+499,General VQA,allava_vflan,-2.6130025,14.024593
+500,General VQA,allava_vflan,-0.81916964,13.724776
+501,General VQA,allava_vflan,-3.6751776,12.906602
+502,General VQA,allava_vflan,-4.2783747,14.749102
+503,General VQA,allava_vflan,-1.5372581,17.109665
+504,General VQA,allava_vflan,-2.0776,15.193711
+505,General VQA,allava_vflan,-4.4844775,15.745609
+506,OCR QA,aokvqa,-5.036949,15.471095
+507,Naive OCR,art,-3.7019806,12.882121
+508,OCR QA,arxivqa,-5.619008,12.679088
+509,OCR QA,arxivqa,-5.1907144,12.827566
+510,OCR QA,arxivqa,-4.9649296,12.824779
+511,OCR QA,arxivqa,-5.311354,12.7939825
+512,OCR QA,arxivqa,-5.0563145,17.333036
+513,OCR QA,arxivqa,-4.8749228,13.736463
+514,OCR QA,arxivqa,-5.161104,13.196506
+515,OCR QA,arxivqa,-5.959082,15.139402
+516,OCR QA,arxivqa,-3.6331475,15.512168
+517,OCR QA,arxivqa,-5.052316,12.907017
+518,OCR QA,bentham,2.1422567,11.893601
+519,OCR QA,blockdiagramcomputerized,-5.2001925,14.390779
+520,OCR QA,blockdiagramhandwritten,-5.0984387,13.780292
+521,General VQA,cambrian(filtered)_processed,-1.1905512,15.165925
+522,General VQA,cambrian(filtered)_processed,-1.751029,9.587367
+523,General VQA,cambrian(filtered)_processed,-1.6406345,9.58576
+524,General VQA,cambrian(filtered)_processed,-2.7925062,13.11975
+525,General VQA,cambrian(filtered)_processed,-6.0342836,15.802432
+526,General VQA,cambrian(filtered)_processed,-1.1502693,16.989122
+527,General VQA,cambrian(filtered)_processed,-2.4158435,14.082861
+528,General VQA,cambrian(filtered)_processed,-1.2166952,13.693319
+529,Naive OCR,captcha,0.22763692,12.495235
+530,Naive OCR,captcha,0.9593981,12.363588
+531,Naive OCR,captcha,0.081198886,12.573855
+532,Naive OCR,captcha,0.36387718,12.624115
+533,Naive OCR,captcha,0.6993159,12.38368
+534,Naive OCR,captcha,0.85992694,12.335283
+535,Naive OCR,captcha,0.008458383,12.324167
+536,Naive OCR,captcha,0.85702163,12.322404
+537,Naive OCR,captcha,0.02061756,12.329512
+538,Naive OCR,captcha,0.3036315,12.321469
+539,Naive OCR,captcha,0.2679477,12.438782
+540,Chart & Table,chart2text,-6.5100636,15.171456
+541,Chart & Table,chart2text,-7.1274424,15.56234
+542,Chart & Table,chartqa,-7.0069585,16.346754
+543,General VQA,chinesememe,-2.0163398,12.233347
+544,General VQA,chinesememe,-0.40541923,14.034706
+545,General VQA,chinesememe,-4.276584,17.076757
+546,General VQA,chinesememe,-2.3813174,14.376326
+547,General VQA,chinesememe,-1.7785631,13.145592
+548,Naive OCR,chrome_writting,1.1499455,13.03082
+549,Mathematics,clevr,-1.6076629,9.8073225
+550,Mathematics,clevr,-1.7101148,10.113159
+551,Mathematics,clevr,-2.0265934,9.784586
+552,Mathematics,clevr,-2.1135507,9.829596
+553,Mathematics,clevr,-1.9405135,9.9880085
+554,Mathematics,clevr,-2.196799,10.16171
+555,Mathematics,clevr,-1.452498,9.7752285
+556,Mathematics,clevr_math,-2.1801212,9.801649
+557,Mathematics,clevr_math,-1.9273381,9.612518
+558,Mathematics,clevr_math,-1.5411426,9.655872
+559,Mathematics,clevr_math,-1.7625996,9.580066
+560,Mathematics,clevr_math,-1.9678849,10.023421
+561,Mathematics,clevr_math,-1.7113199,9.879975
+562,Mathematics,clevr_math,-1.484115,9.620203
+563,Mathematics,clevr_math(mathv360k),-1.9961169,9.7263975
+564,Captioning & Knowledge,coco_colors,-1.3454305,15.751123
+565,Captioning & Knowledge,coco_colors,-3.144214,13.285055
+566,Captioning & Knowledge,coco_colors,-2.0505533,15.2906065
+567,Captioning & Knowledge,coco_colors,-2.5754852,13.428278
+568,Captioning & Knowledge,coco_colors,-2.0880659,13.858809
+569,Captioning & Knowledge,coco_colors,-1.2876897,11.391866
+570,Captioning & Knowledge,coco_colors,-2.963714,13.377311
+571,Captioning & Knowledge,coco_colors,-6.594825,14.224752
+572,Captioning & Knowledge,coco_colors,-3.8137763,12.6804085
+573,Captioning & Knowledge,coco_colors,-2.0886075,11.634719
+574,Captioning & Knowledge,coco_colors,-1.3111572,14.943168
+575,General VQA,cocoqa,-3.267713,13.048398
+576,General VQA,cocoqa,-0.013037341,14.5589695
+577,General VQA,cocoqa,-1.7347096,11.373774
+578,General VQA,cocoqa,-2.1878505,15.053922
+579,Naive OCR,cocotext,-1.1052303,13.150914
+580,Naive OCR,ctw,-0.86423486,16.745403
+581,Naive OCR,ctw,-1.4000857,13.775461
+582,General VQA,datik,-6.2355814,14.348469
+583,General VQA,datik,-5.5889235,11.836506
+584,General VQA,datik,-6.0488763,14.155444
+585,General VQA,datik,-5.6327734,16.707447
+586,General VQA,datik,-4.9528985,11.5223055
+587,General VQA,datik,-5.85656,13.457073
+588,General VQA,datik,-5.373902,12.447471
+589,General VQA,datik,-6.1315455,12.849513
+590,General VQA,datik,-4.7576,12.136585
+591,General VQA,datik,-4.7864375,14.507768
+592,General VQA,datik,-4.4987044,13.338192
+593,General VQA,datik,-5.1533957,12.187311
+594,General VQA,datik,-5.552182,12.109758
+595,General VQA,datik,-5.2981977,12.0126
+596,General VQA,datik,-5.423341,12.789884
+597,General VQA,datik,-5.3969746,12.0191555
+598,General VQA,datik,-5.676181,12.3539715
+599,General VQA,datik,-3.8894515,12.648885
+600,General VQA,datik,-4.9521174,11.567204
+601,General VQA,datik,-4.827991,11.952656
+602,General VQA,datik,-5.789037,12.559905
+603,General VQA,datik,-4.80648,14.409976
+604,General VQA,datikz,-5.4925127,12.3455515
+605,General VQA,datikz,-0.48186696,16.16223
+606,General VQA,datikz,-4.5972114,12.744012
+607,General VQA,datikz,-5.7660947,14.133665
+608,Captioning & Knowledge,densefusion_1m,-3.3065581,15.509387
+609,Captioning & Knowledge,densefusion_1m,-3.5640764,14.870224
+610,Captioning & Knowledge,densefusion_1m,-5.4659333,17.194544
+611,Captioning & Knowledge,densefusion_1m,-0.13317387,13.016425
+612,Captioning & Knowledge,densefusion_1m,-3.348509,13.680758
+613,Captioning & Knowledge,densefusion_1m,-3.6600456,13.614563
+614,Captioning & Knowledge,densefusion_1m,-2.5580869,15.025556
+615,Captioning & Knowledge,densefusion_1m,-4.6975946,13.420876
+616,Captioning & Knowledge,densefusion_1m,-5.1021976,15.629852
+617,Captioning & Knowledge,densefusion_1m,-3.3262618,16.307451
+618,Captioning & Knowledge,densefusion_1m,0.5171699,15.874338
+619,Captioning & Knowledge,densefusion_1m,-3.1347923,13.879251
+620,Captioning & Knowledge,densefusion_1m,-3.0242152,14.236134
+621,Captioning & Knowledge,densefusion_1m,-2.8946476,15.891527
+622,Captioning & Knowledge,densefusion_1m,-4.222332,15.951677
+623,Captioning & Knowledge,densefusion_1m,-2.3120508,13.763496
+624,Captioning & Knowledge,densefusion_1m,-3.61744,15.599881
+625,Captioning & Knowledge,densefusion_1m,-2.0494604,17.18496
+626,Captioning & Knowledge,densefusion_1m,-3.942528,13.995651
+627,Captioning & Knowledge,densefusion_1m,-2.0762098,16.159477
+628,Captioning & Knowledge,densefusion_1m,-3.7376199,15.6758
+629,Captioning & Knowledge,densefusion_1m,-3.4259229,14.399123
+630,Captioning & Knowledge,densefusion_1m,-2.5529807,16.250513
+631,Captioning & Knowledge,densefusion_1m,-2.6858516,14.679809
+632,Captioning & Knowledge,densefusion_1m,-2.5089886,15.47572
+633,Captioning & Knowledge,densefusion_1m,-2.120121,16.392109
+634,Captioning & Knowledge,densefusion_1m,-1.3603035,13.391585
+635,Captioning & Knowledge,densefusion_1m,-2.7239373,14.706802
+636,Captioning & Knowledge,densefusion_1m,-0.07262699,14.361336
+637,Captioning & Knowledge,densefusion_1m,-1.8037416,15.877097
+638,Captioning & Knowledge,densefusion_1m,-2.941668,16.743402
+639,Captioning & Knowledge,densefusion_1m,-4.2678475,14.512948
+640,Captioning & Knowledge,densefusion_1m,-4.041674,16.051338
+641,Captioning & Knowledge,densefusion_1m,-2.5236688,14.999438
+642,Captioning & Knowledge,densefusion_1m,-2.026976,15.456689
+643,Captioning & Knowledge,densefusion_1m,-2.0059025,11.478699
+644,Captioning & Knowledge,densefusion_1m,-4.50681,14.990721
+645,Captioning & Knowledge,densefusion_1m,-3.3491118,17.68867
+646,Captioning & Knowledge,densefusion_1m,-4.277411,16.145494
+647,Captioning & Knowledge,densefusion_1m,0.52266985,13.212613
+648,Captioning & Knowledge,densefusion_1m,-3.4547403,16.325184
+649,Captioning & Knowledge,densefusion_1m,-0.90948886,14.223853
+650,Captioning & Knowledge,densefusion_1m,-2.4410052,14.590959
+651,Captioning & Knowledge,densefusion_1m,-3.6672225,16.200733
+652,Captioning & Knowledge,densefusion_1m,-0.13326755,16.51813
+653,Captioning & Knowledge,densefusion_1m,-2.5357149,12.886249
+654,Captioning & Knowledge,densefusion_1m,-4.3852863,17.96995
+655,Captioning & Knowledge,densefusion_1m,-1.0622027,14.245199
+656,Captioning & Knowledge,densefusion_1m,-1.5756062,16.210028
+657,Captioning & Knowledge,densefusion_1m,-2.3257315,16.436813
+658,Captioning & Knowledge,densefusion_1m,-2.310822,17.15486
+659,Captioning & Knowledge,densefusion_1m,-4.7023654,13.6322365
+660,Captioning & Knowledge,densefusion_1m,-2.9235616,13.916572
+661,Captioning & Knowledge,densefusion_1m,-2.130736,13.8982935
+662,Captioning & Knowledge,densefusion_1m,-3.9695623,15.76031
+663,Captioning & Knowledge,densefusion_1m,-3.0472314,15.9308605
+664,Captioning & Knowledge,densefusion_1m,-2.0730689,15.980832
+665,Captioning & Knowledge,densefusion_1m,-3.242564,15.050029
+666,Captioning & Knowledge,densefusion_1m,-1.7463053,15.30607
+667,Captioning & Knowledge,densefusion_1m,-3.2756486,14.984592
+668,Captioning & Knowledge,densefusion_1m,-2.1766455,13.659809
+669,Captioning & Knowledge,densefusion_1m,-3.246647,16.380726
+670,Captioning & Knowledge,densefusion_1m,-0.9537033,11.664696
+671,Captioning & Knowledge,densefusion_1m,0.92756504,11.384646
+672,Captioning & Knowledge,densefusion_1m,-1.2184327,14.4949255
+673,Captioning & Knowledge,densefusion_1m,-2.1183088,14.774638
+674,Captioning & Knowledge,densefusion_1m,1.1054575,11.77428
+675,Captioning & Knowledge,densefusion_1m,-2.2946916,18.594072
+676,Captioning & Knowledge,densefusion_1m,-2.1221104,15.252288
+677,Captioning & Knowledge,densefusion_1m,-2.660571,15.598898
+678,Captioning & Knowledge,densefusion_1m,-2.3529189,14.205819
+679,Captioning & Knowledge,densefusion_1m,-2.0088046,15.155662
+680,Captioning & Knowledge,densefusion_1m,-2.1137953,14.337944
+681,Captioning & Knowledge,densefusion_1m,-2.5796294,15.830268
+682,Captioning & Knowledge,densefusion_1m,-1.0930537,14.505688
+683,Captioning & Knowledge,densefusion_1m,-1.76625,10.75198
+684,Captioning & Knowledge,densefusion_1m,-2.8316212,17.15655
+685,Captioning & Knowledge,densefusion_1m,-2.3657508,14.273163
+686,Captioning & Knowledge,densefusion_1m,-3.210463,15.3637295
+687,Captioning & Knowledge,densefusion_1m,-2.3319025,14.658825
+688,Captioning & Knowledge,densefusion_1m,-4.2783475,18.116081
+689,Captioning & Knowledge,densefusion_1m,-3.0389729,11.39726
+690,Captioning & Knowledge,densefusion_1m,-4.6191506,13.757706
+691,Captioning & Knowledge,densefusion_1m,-3.2199864,15.038432
+692,Captioning & Knowledge,densefusion_1m,-4.806803,16.297205
+693,Captioning & Knowledge,densefusion_1m,-2.5765905,15.238693
+694,Captioning & Knowledge,densefusion_1m,-2.555151,14.315961
+695,Captioning & Knowledge,densefusion_1m,-3.3783765,16.171057
+696,Captioning & Knowledge,densefusion_1m,-2.404972,15.834854
+697,Captioning & Knowledge,densefusion_1m,-1.7184036,16.265429
+698,Captioning & Knowledge,densefusion_1m,-3.0274854,15.919806
+699,Captioning & Knowledge,densefusion_1m,-3.5488534,15.739376
+700,Captioning & Knowledge,densefusion_1m,0.17561264,14.762424
+701,Captioning & Knowledge,densefusion_1m,-2.3952467,14.774095
+702,Captioning & Knowledge,densefusion_1m,-2.4548035,17.044546
+703,Captioning & Knowledge,densefusion_1m,-2.5112517,13.688839
+704,Captioning & Knowledge,densefusion_1m,-4.3332334,15.496156
+705,Captioning & Knowledge,densefusion_1m,-3.8103876,16.509098
+706,Captioning & Knowledge,densefusion_1m,-2.2640226,13.017393
+707,Captioning & Knowledge,densefusion_1m,1.3102521,15.414042
+708,Captioning & Knowledge,densefusion_1m,-3.2106593,14.903614
+709,Captioning & Knowledge,densefusion_1m,-5.1128235,15.946321
+710,Captioning & Knowledge,densefusion_1m,-3.0940828,15.633901
+711,Captioning & Knowledge,densefusion_1m,0.32221013,14.868876
+712,Captioning & Knowledge,densefusion_1m,-3.7157872,14.398144
+713,OCR QA,diagram_image_to_text,-4.866783,13.359544
+714,OCR QA,docvqa,-3.0494528,18.62894
+715,General VQA,drivelm,-1.0703018,12.464253
+716,Chart & Table,dvqa,-6.242144,15.664062
+717,Chart & Table,dvqa,-6.5798836,14.925664
+718,Chart & Table,dvqa,-5.6856976,14.456819
+719,Chart & Table,dvqa,-6.4766316,14.437862
+720,Chart & Table,dvqa,-6.296074,14.4348545
+721,Chart & Table,dvqa,-6.6247945,15.171215
+722,Chart & Table,dvqa,-5.974128,16.190802
+723,Chart & Table,dvqa,-6.1348386,14.469602
+724,Chart & Table,dvqa,-6.1982417,15.850152
+725,Chart & Table,dvqa,-6.021088,14.488718
+726,Chart & Table,dvqa,-6.0421634,15.962641
+727,Chart & Table,dvqa,-6.690761,15.004449
+728,Chart & Table,dvqa,-5.915554,15.966814
+729,Chart & Table,dvqa,-6.7349772,14.572269
+730,Chart & Table,dvqa,-6.297472,16.45757
+731,Chart & Table,dvqa,-5.479183,16.141006
+732,Chart & Table,dvqa,-6.644602,14.768048
+733,Chart & Table,dvqa,-6.237617,14.749125
+734,Chart & Table,dvqa,-6.134201,14.719306
+735,Chart & Table,dvqa,-6.7095127,14.807197
+736,OCR QA,est_vqa,-2.8028898,13.487987
+737,Captioning & Knowledge,face_emotion,-1.7859892,12.350736
+738,Chart & Table,figureqa,-4.7813435,17.928652
+739,Chart & Table,figureqa,-6.784873,15.886794
+740,Chart & Table,figureqa,-5.3147545,18.12775
+741,Chart & Table,figureqa,-5.3968844,12.893658
+742,Chart & Table,figureqa,-6.6792145,15.061978
+743,Chart & Table,figureqa,-5.2459846,13.071614
+744,Chart & Table,figureqa,-6.6299553,14.906391
+745,Chart & Table,figureqa,-6.0355377,13.285192
+746,Chart & Table,figureqa,-5.4043474,12.8902235
+747,Chart & Table,figureqa,-5.23914,12.990279
+748,Chart & Table,figureqa(mathv360k),-5.8802824,13.058701
+749,Chart & Table,finqa,-4.88894,16.396334
+750,Naive OCR,funsd,-1.756222,16.794167
+751,Mathematics,geo170k(align),-3.8987415,12.288033
+752,Mathematics,geo170k(align),-4.8021193,12.065388
+753,Mathematics,geo170k(align),-4.8182483,11.813766
+754,Mathematics,geo170k(qa),-4.8915815,11.612733
+755,Mathematics,geo3k,-4.682466,12.288988
+756,Mathematics,geometry3k(mathv360k),-4.7258954,13.311209
+757,Mathematics,geomverse,-6.0546713,17.217825
+758,Mathematics,geoqa+(mathv360k),-4.881611,12.036243
+759,Mathematics,geos(mathv360k),-4.5555034,11.766478
+760,Captioning & Knowledge,google_landmarks,0.20172866,13.901081
+761,Captioning & Knowledge,google_landmarks,-2.6422932,12.871565
+762,Captioning & Knowledge,google_landmarks,-3.8555532,17.991074
+763,Captioning & Knowledge,google_landmarks,0.040216874,12.912093
+764,Captioning & Knowledge,google_landmarks,-1.0884169,12.750918
+765,Captioning & Knowledge,google_landmarks,-1.1757756,15.603528
+766,Captioning & Knowledge,google_landmarks,-3.2434611,14.709588
+767,Captioning & Knowledge,google_landmarks,-4.3859215,14.216998
+768,Captioning & Knowledge,google_landmarks,-3.8517127,12.949343
+769,Captioning & Knowledge,google_landmarks,-5.0439453,16.073694
+770,Captioning & Knowledge,google_landmarks,-0.98564696,15.400732
+771,Captioning & Knowledge,google_landmarks,-0.69483256,14.727176
+772,Captioning & Knowledge,google_landmarks,-0.72640425,15.866496
+773,Captioning & Knowledge,google_landmarks,-2.2906032,15.273375
+774,Captioning & Knowledge,google_landmarks,-0.72795415,11.727866
+775,Captioning & Knowledge,google_landmarks,1.3738796,12.26725
+776,Captioning & Knowledge,google_landmarks,-3.5594008,14.2105465
+777,Captioning & Knowledge,google_landmarks,-1.289138,15.407301
+778,Captioning & Knowledge,google_landmarks,-0.17793657,12.992344
+779,Captioning & Knowledge,google_landmarks,-2.1144824,13.824165
+780,Captioning & Knowledge,google_landmarks,-3.4853973,13.422749
+781,Captioning & Knowledge,google_landmarks,-4.9478903,17.307014
+782,Captioning & Knowledge,google_landmarks,-2.553987,16.698412
+783,Captioning & Knowledge,google_landmarks,-1.5745763,14.7828
+784,Captioning & Knowledge,google_landmarks,-4.3067718,13.16484
+785,Captioning & Knowledge,google_landmarks,-1.6564403,12.325743
+786,Captioning & Knowledge,google_landmarks,-3.4131317,15.211848
+787,Captioning & Knowledge,google_landmarks,-3.965186,15.60332
+788,Captioning & Knowledge,google_landmarks,-3.1230712,14.69487
+789,Grounding & Counting,groundui,-2.711454,16.711662
+790,OCR QA,handwriting_forms,0.82387245,14.322479
+791,General VQA,hateful_memes,-2.0540226,14.267113
+792,Chart & Table,hitab,-5.0266867,16.74949
+793,Naive OCR,hme100k,1.8126867,12.138864
+794,Naive OCR,hme100k,2.1684346,12.54377
+795,Naive OCR,hme100k,1.6003158,12.643909
+796,Naive OCR,hme100k,-0.46795946,11.392594
+797,Naive OCR,hme100k,1.8394164,12.640122
+798,Naive OCR,hme100k,1.6946428,12.5427475
+799,Naive OCR,hme100k,2.0096955,12.31766
+800,Naive OCR,hw_squad,-3.5496185,13.323637
+801,Naive OCR,hw_squad,-3.3809865,12.351537
+802,Naive OCR,iam,1.6615646,13.043853
+803,General VQA,iconqa,-6.022064,15.315262
+804,General VQA,iconqa,-4.49422,13.14323
+805,General VQA,iconqa(mathv360k),1.9024354,12.974247
+806,General VQA,iconqa(mathv360k),-3.9231687,13.202329
+807,General VQA,idk,-2.6233222,12.832308
+808,Naive OCR,iiit5k,0.3202509,12.139457
+809,Captioning & Knowledge,image_textualization(filtered),-3.3486886,13.058075
+810,Captioning & Knowledge,image_textualization(filtered),-4.822362,14.919619
+811,Captioning & Knowledge,image_textualization(filtered),-2.5731385,16.531271
+812,Captioning & Knowledge,image_textualization(filtered),-1.4716326,13.933705
+813,Captioning & Knowledge,image_textualization(filtered),-2.3679855,15.059535
+814,Captioning & Knowledge,image_textualization(filtered),0.6983169,14.216063
+815,Captioning & Knowledge,image_textualization(filtered),-3.3752155,14.5489
+816,Captioning & Knowledge,image_textualization(filtered),-1.2714372,12.940445
+817,Captioning & Knowledge,image_textualization(filtered),-2.9996772,16.013012
+818,Naive OCR,imgur5k,-0.87933296,11.392932
+819,General VQA,indoor_qa,-3.8503525,14.082553
+820,OCR QA,infographic(gpt4v),-3.1881175,15.614101
+821,OCR QA,infographic_vqa,-2.1972423,17.634672
+822,OCR QA,infographic_vqa_llava_format,-3.4816718,15.361047
+823,Mathematics,intergps,-4.2479253,11.915455
+824,OCR QA,invoices_receipts,-3.0610104,10.959195
+825,Naive OCR,k12_printing,1.1667885,11.798061
+826,Naive OCR,k12_printing,1.2854391,11.182922
+827,Naive OCR,k12_printing,2.1556146,11.567906
+828,Naive OCR,k12_printing,2.0654805,11.784395
+829,Naive OCR,k12_printing,1.7834607,10.793035
+830,Naive OCR,k12_printing,1.0891058,12.097322
+831,Naive OCR,k12_printing,1.8238891,10.8158045
+832,Naive OCR,k12_printing,1.3869863,11.935883
+833,Naive OCR,k12_printing,2.502718,11.500277
+834,Naive OCR,k12_printing,1.9121165,10.788772
+835,Naive OCR,k12_printing,2.2414637,11.012823
+836,Naive OCR,k12_printing,1.7801155,11.367645
+837,Naive OCR,k12_printing,1.6598969,11.890178
+838,Naive OCR,k12_printing,1.0697793,11.743265
+839,Naive OCR,k12_printing,1.895159,13.251504
+840,Naive OCR,k12_printing,1.4264286,10.895116
+841,Naive OCR,k12_printing,1.7905657,10.904459
+842,Naive OCR,k12_printing,1.8476293,11.29355
+843,Naive OCR,k12_printing,1.7207872,10.815492
+844,Naive OCR,k12_printing,1.653414,11.016292
+845,Naive OCR,k12_printing,2.4655066,11.557249
+846,Naive OCR,k12_printing,1.4819993,12.0113535
+847,Naive OCR,k12_printing,1.3671023,10.840212
+848,Naive OCR,k12_printing,2.459251,11.674356
+849,Naive OCR,k12_printing,2.1497476,11.000881
+850,Captioning & Knowledge,laion_gpt4v,-4.1161394,12.105045
+851,Naive OCR,latex_handwritten,-0.8454308,11.3400135
+852,Naive OCR,latex_handwritten,1.4271748,13.32655
+853,Naive OCR,latex_handwritten,-0.41473493,11.451408
+854,Naive OCR,latexformulas,2.516351,13.928785
+855,Naive OCR,latexformulas,2.602297,14.593694
+856,Naive OCR,latexformulas,1.8524942,15.453784
+857,Naive OCR,latexformulas,2.3205998,14.35518
+858,Naive OCR,latexformulas,1.8644116,14.885029
+859,Naive OCR,latexformulas,2.7975779,14.18772
+860,Naive OCR,latexformulas,2.3942838,13.322677
+861,Naive OCR,latexformulas,1.4445506,14.220874
+862,Naive OCR,latexformulas,2.2836862,11.39136
+863,Naive OCR,latexformulas,2.6057982,13.961386
+864,Naive OCR,latexformulas,1.9961302,13.434808
+865,Naive OCR,latexformulas,1.8952883,14.131218
+866,Naive OCR,latexformulas,2.255458,14.974293
+867,Naive OCR,latexformulas,2.6445436,14.058138
+868,Naive OCR,latexformulas,1.6173706,15.434319
+869,Naive OCR,latexformulas,2.3632889,13.453909
+870,Naive OCR,latexformulas,2.2996132,15.094052
+871,Naive OCR,latexformulas,2.5076869,11.733104
+872,Naive OCR,latexformulas,1.5564226,15.268579
+873,Naive OCR,latexformulas,2.4036155,13.5982485
+874,Naive OCR,latexformulas,1.5522492,14.003594
+875,Naive OCR,latexformulas,2.1294045,15.243942
+876,Naive OCR,latexformulas,2.6221385,14.464621
+877,Naive OCR,latexformulas,2.0067592,15.298331
+878,Naive OCR,latexformulas,1.2816901,14.289165
+879,Naive OCR,latexformulas,1.719467,14.07952
+880,Naive OCR,latexformulas,2.5884748,14.228993
+881,Naive OCR,latexformulas,2.8904116,14.023879
+882,Naive OCR,latexformulas,2.2152932,14.760879
+883,Naive OCR,latexformulas,1.5691396,13.88199
+884,Naive OCR,latexformulas,2.801726,14.41394
+885,Naive OCR,latexformulas,2.3628526,14.097104
+886,Naive OCR,latexformulas,2.160742,13.56287
+887,Naive OCR,latexformulas,2.2690992,14.015383
+888,Naive OCR,latexformulas,2.4847662,14.191013
+889,Naive OCR,latexformulas,2.3792138,13.203296
+890,Naive OCR,latexformulas,2.4925876,13.499819
+891,Naive OCR,latexformulas,2.7798162,13.650808
+892,Naive OCR,latexformulas,2.276965,14.88868
+893,Naive OCR,latexformulas,2.6273801,14.053435
+894,Naive OCR,latexformulas,1.6203766,14.501678
+895,Naive OCR,latexformulas,2.1433609,13.745667
+896,Naive OCR,latexformulas,2.608872,14.691044
+897,Naive OCR,latexformulas,2.4453807,13.2606735
+898,Naive OCR,latexformulas,2.5142748,15.041762
+899,Naive OCR,latexformulas,2.227619,13.409311
+900,Naive OCR,latexformulas,1.5289643,14.450167
+901,Naive OCR,latexformulas,2.1774447,14.9880495
+902,Naive OCR,latexformulas,2.1724415,14.356275
+903,Naive OCR,latexformulas,2.920621,14.340328
+904,Naive OCR,latexformulas,2.3055284,14.393307
+905,Naive OCR,latexformulas,2.9679034,13.975026
+906,Naive OCR,latexformulas,2.6873116,13.886729
+907,Naive OCR,latexformulas,2.2397995,14.439811
+908,Naive OCR,latexformulas,2.0753183,13.854256
+909,General VQA,llavar_gpt4_20k,-1.7171866,15.822935
+910,General VQA,lnqa,-1.8064234,15.756924
+911,General VQA,lnqa,-0.7592474,13.04315
+912,General VQA,lnqa,-1.6349295,13.438519
+913,General VQA,lnqa,-5.131853,15.128572
+914,General VQA,lnqa,-0.48803896,15.147016
+915,General VQA,lnqa,-1.0575602,13.713107
+916,General VQA,lnqa,-1.048233,12.52339
+917,General VQA,lnqa,-1.0816207,15.113961
+918,General VQA,lnqa,-0.9096639,14.210454
+919,General VQA,lnqa,-2.9719315,13.746634
+920,General VQA,lnqa,-1.5493317,14.457064
+921,General VQA,lnqa,-2.6358893,12.372189
+922,General VQA,lnqa,-2.4512897,14.645212
+923,General VQA,lnqa,-1.624204,15.008271
+924,General VQA,lnqa,-1.0908378,14.259123
+925,General VQA,lnqa,-1.3196355,16.124268
+926,General VQA,lnqa,-1.3070955,12.0567465
+927,General VQA,lnqa,-2.239633,12.991782
+928,General VQA,lnqa,-2.8251464,12.3505125
+929,General VQA,lnqa,-2.2722502,14.346933
+930,General VQA,lnqa,-0.690192,14.2005005
+931,General VQA,lnqa,-2.6837037,13.142105
+932,General VQA,lnqa,-3.254341,17.457916
+933,General VQA,lnqa,-4.3692985,15.596879
+934,General VQA,lnqa,-2.5166562,13.018639
+935,General VQA,lnqa,-5.846177,16.989979
+936,General VQA,lnqa,-1.4032556,13.183372
+937,General VQA,lnqa,-1.994133,13.825772
+938,General VQA,lnqa,-4.2416644,12.5169325
+939,General VQA,lnqa,-3.1771703,13.739716
+940,Captioning & Knowledge,localized_narratives,-3.644063,12.387532
+941,Captioning & Knowledge,localized_narratives,-1.3647183,15.610713
+942,Captioning & Knowledge,localized_narratives,-1.9175681,12.7565
+943,Captioning & Knowledge,localized_narratives,-2.3589046,13.145044
+944,Captioning & Knowledge,localized_narratives,-3.6379864,13.342935
+945,Captioning & Knowledge,localized_narratives,-0.81609595,17.167885
+946,Captioning & Knowledge,localized_narratives,-2.5721328,13.474737
+947,Captioning & Knowledge,localized_narratives,-5.2345195,13.409462
+948,Captioning & Knowledge,localized_narratives,-0.18872811,13.813309
+949,Captioning & Knowledge,localized_narratives,0.12778817,14.1336355
+950,Captioning & Knowledge,localized_narratives,-3.2016358,11.858263
+951,Captioning & Knowledge,localized_narratives,-0.6901099,12.549499
+952,Captioning & Knowledge,localized_narratives,-1.4131317,13.318709
+953,Captioning & Knowledge,localized_narratives,-3.667536,11.874386
+954,Captioning & Knowledge,localized_narratives,-1.0671754,13.6831
+955,Captioning & Knowledge,localized_narratives,-0.7765497,15.164078
+956,Captioning & Knowledge,localized_narratives,-2.1577964,13.48058
+957,Captioning & Knowledge,localized_narratives,-1.4175575,14.640538
+958,Captioning & Knowledge,localized_narratives,-2.1317937,12.812771
+959,Chart & Table,lrv_chart,-6.273305,13.3217535
+960,General VQA,lrv_normal(filtered),-0.76939243,12.967264
+961,General VQA,lvis_instruct4v,-4.73657,14.83389
+962,General VQA,lvis_instruct4v,-2.4940844,12.145509
+963,General VQA,lvis_instruct4v,-0.9268688,12.656746
+964,General VQA,lvis_instruct4v,-4.7394056,17.78486
+965,General VQA,lvis_instruct4v,-0.3814387,14.323441
+966,General VQA,lvis_instruct4v,-0.53778,15.835095
+967,General VQA,lvis_instruct4v,-2.3239827,16.329975
+968,General VQA,lvis_instruct4v,-2.2501986,15.653687
+969,General VQA,lvis_instruct4v,-2.7152863,14.495052
+970,General VQA,lvis_instruct4v,-4.2499375,15.668114
+971,General VQA,lvis_instruct4v,-0.5543433,12.862728
+972,General VQA,lvis_instruct4v,-2.4506278,13.771455
+973,General VQA,lvis_instruct4v,-1.1804779,14.664362
+974,General VQA,lvis_instruct4v,-2.7578933,13.63672
+975,General VQA,lvis_instruct4v,-4.091878,14.817504
+976,General VQA,lvis_instruct4v,-3.3705537,16.399845
+977,General VQA,lvis_instruct4v,-2.8542416,13.804453
+978,General VQA,lvis_instruct4v,-2.0456831,14.836426
+979,General VQA,lvis_instruct4v,-0.72630984,13.661054
+980,General VQA,lvis_instruct4v,-0.3347137,14.390195
+981,General VQA,lvis_instruct4v,-2.4582636,12.975029
+982,General VQA,lvis_instruct4v,-0.99313223,15.641524
+983,OCR QA,mapqa,-5.043743,15.640015
+984,OCR QA,mapqa,-4.9534144,15.529422
+985,OCR QA,mapqa,-5.221481,15.934826
+986,OCR QA,mapqa(mathv360k),-4.907941,15.241494
+987,Naive OCR,maptext,-2.5046973,13.238817
+988,Naive OCR,mathwriting-google,0.54370576,11.116353
+989,Naive OCR,mathwriting-google,-0.6979461,10.774328
+990,Naive OCR,mathwriting-google,-0.5098082,10.840544
+991,Naive OCR,mathwriting-google,-1.0690495,11.180002
+992,Naive OCR,mathwriting-google,-0.22856215,11.109361
+993,Naive OCR,mathwriting-google,-0.8797214,10.695546
+994,Naive OCR,mathwriting-google,-0.33506313,11.12741
+995,Naive OCR,mathwriting-google,-0.96779305,10.750852
+996,Naive OCR,mathwriting-google,-0.6270284,10.77479
+997,Naive OCR,mathwriting-google,1.405327,12.681331
+998,Naive OCR,mathwriting-google,-0.70240825,10.842642
+999,Naive OCR,mathwriting-google,0.07038674,11.0433035
+1000,Naive OCR,mathwriting-google,0.0018071033,10.844643
+1001,Naive OCR,mathwriting-google,-1.5012355,11.031652
+1002,Naive OCR,mathwriting-google,-1.2133697,10.769429
+1003,Naive OCR,mathwriting-google,-0.1749257,10.799909
+1004,Naive OCR,mathwriting-google,-1.3969884,11.000927
+1005,Naive OCR,mathwriting-google,-0.74503976,10.906645
+1006,Naive OCR,mathwriting-google,0.51168174,12.002008
+1007,Naive OCR,mathwriting-google,-1.1593268,11.070854
+1008,Naive OCR,mathwriting-google,-1.0289783,10.887972
+1009,Naive OCR,mathwriting-google,-0.6220506,10.946476
+1010,Naive OCR,mathwriting-google,-0.07094961,10.657791
+1011,Naive OCR,mathwriting-google,-0.5101048,11.075313
+1012,Naive OCR,mathwriting-google,-0.8554856,11.088044
+1013,Naive OCR,mathwriting-google,-0.24182698,10.703694
+1014,Naive OCR,mathwriting-google,-0.07323343,10.776095
+1015,Naive OCR,mathwriting-google,1.0851828,13.895364
+1016,Naive OCR,mathwriting-google,-1.0423876,10.914542
+1017,Naive OCR,mathwriting-google,-0.3726632,10.710663
+1018,Mathematics,mavis_math_metagen,1.0100632,15.34724
+1019,Mathematics,mavis_math_metagen,-4.910757,12.873553
+1020,Mathematics,mavis_math_metagen,-5.5636463,15.337041
+1021,Mathematics,mavis_math_metagen,-4.3060884,12.206949
+1022,Mathematics,mavis_math_metagen,-6.2341475,12.973666
+1023,Mathematics,mavis_math_metagen,-4.9193754,12.852338
+1024,Mathematics,mavis_math_metagen,-6.082518,12.738337
+1025,Mathematics,mavis_math_metagen,-4.770181,11.88174
+1026,Mathematics,mavis_math_rule_geo,-5.0410986,11.44003
+1027,Mathematics,mavis_math_rule_geo,-5.2008615,11.959395
+1028,Mathematics,mavis_math_rule_geo,-5.2958803,11.4623575
+1029,Mathematics,mavis_math_rule_geo,-5.199357,11.469481
+1030,Mathematics,mavis_math_rule_geo,-5.304515,11.725547
+1031,Mathematics,mavis_math_rule_geo,-5.411612,11.518962
+1032,Mathematics,mavis_math_rule_geo,-5.5468063,12.109462
+1033,Mathematics,mavis_math_rule_geo,-5.3459396,11.493222
+1034,Mathematics,mavis_math_rule_geo,-5.138787,11.569318
+1035,Naive OCR,memotion,-2.7186327,14.725555
+1036,General VQA,mimic_cgd,-2.8202019,14.28582
+1037,General VQA,mimic_cgd,-1.5461861,12.22294
+1038,General VQA,mimic_cgd,0.12196187,12.688118
+1039,General VQA,mimic_cgd,-0.08615107,14.6829195
+1040,General VQA,mimic_cgd,-2.621902,12.118658
+1041,General VQA,mimic_cgd,-0.49144223,13.316888
+1042,General VQA,mimic_cgd,-1.8795115,12.5145
+1043,Chart & Table,mmc_instruct,-4.60311,14.285842
+1044,Chart & Table,mmc_instruct,-5.03259,14.211798
+1045,Chart & Table,mmc_instruct,-5.129204,14.193633
+1046,Chart & Table,mmc_instruct,-5.249871,13.62388
+1047,Chart & Table,mmc_instruct,-5.07639,13.964945
+1048,Chart & Table,mmc_instruct,-4.959758,14.094965
+1049,Chart & Table,mmc_instruct,-5.2076383,15.217213
+1050,Chart & Table,mmc_instruct,-5.094092,14.4442625
+1051,Chart & Table,mmc_instruct,-5.0246067,14.048986
+1052,Chart & Table,mmc_instruct,-4.935757,14.302881
+1053,Chart & Table,mmc_instruct,-4.887438,14.252479
+1054,Chart & Table,mmc_instruct,-4.9374943,14.805857
+1055,Chart & Table,mmc_instruct,-5.093808,14.461699
+1056,Chart & Table,mmc_instruct,-5.088552,13.918857
+1057,Chart & Table,mmc_instruct,-5.2426095,13.930322
+1058,Chart & Table,mmc_instruct,-4.677662,15.052288
+1059,General VQA,mmevol,-0.55660343,13.834291
+1060,General VQA,mmevol,-3.7441423,14.768758
+1061,General VQA,mmevol,-6.062779,13.975012
+1062,General VQA,mmevol,-0.42439243,12.57549
+1063,General VQA,mmevol,-6.9340534,15.120116
+1064,General VQA,mmevol,-1.6160824,13.259088
+1065,General VQA,mmevol,-6.0852633,16.039688
+1066,General VQA,mmevol,-0.6690507,13.844877
+1067,General VQA,mmevol,-1.821449,13.970612
+1068,General VQA,mmevol,-4.6309824,14.807651
+1069,General VQA,mmevol,-0.08206473,13.277907
+1070,General VQA,mmevol,-1.8077364,9.824532
+1071,General VQA,mmevol,-4.016271,13.153515
+1072,General VQA,mmevol,-3.7768297,13.251549
+1073,General VQA,mmevol,-6.6345544,15.174003
+1074,General VQA,mmevol,-2.886789,17.481289
+1075,General VQA,mmra,-1.6938553,13.186094
+1076,OCR QA,mmsoc_memotion,-1.6869607,15.576091
+1077,Chart & Table,multihiertt,-4.9029856,16.678263
+1078,General VQA,nlvr2,-0.8234755,15.323158
+1079,General VQA,nlvr2,-2.4119058,16.114187
+1080,General VQA,nlvr2,-3.0032072,13.519479
+1081,General VQA,nlvr2,-1.6724621,12.600614
+1082,General VQA,nlvr2,-3.44456,17.09188
+1083,Grounding & Counting,objects365_qa,-2.7175019,15.437818
+1084,Grounding & Counting,objects365_qa,-2.9432955,13.854029
+1085,Grounding & Counting,objects365_qa,-2.246823,12.408831
+1086,Grounding & Counting,objects365_qa,-1.8299878,14.313221
+1087,Grounding & Counting,objects365_qa,-1.1903279,13.655024
+1088,Grounding & Counting,objects365_qa,-1.8415388,16.28468
+1089,Grounding & Counting,objects365_qa,-2.4823506,12.347847
+1090,Grounding & Counting,objects365_qa,-1.7189224,13.059086
+1091,Grounding & Counting,objects365_qa,-2.0790694,13.356193
+1092,Grounding & Counting,objects365_qa,-1.1196587,14.833971
+1093,Grounding & Counting,objects365_qa,-1.4470162,13.001493
+1094,Grounding & Counting,objects365_qa,-1.028878,14.654299
+1095,Grounding & Counting,objects365_qa,-3.057437,13.19594
+1096,Grounding & Counting,objects365_qa,-0.89064616,14.040172
+1097,Grounding & Counting,objects365_qa,-3.4078298,14.674618
+1098,Grounding & Counting,objects365_qa,-2.314725,13.954356
+1099,Grounding & Counting,objects365_qa,-2.4728005,14.086107
+1100,Grounding & Counting,objects365_qa,-1.9457196,13.709636
+1101,Grounding & Counting,objects365_qa,0.7787471,13.544907
+1102,Grounding & Counting,objects365_qa,-2.9857464,14.147222
+1103,Grounding & Counting,objects365_qa,-0.37216175,13.651664
+1104,Grounding & Counting,objects365_qa,-1.4796355,11.993102
+1105,Grounding & Counting,objects365_qa,-1.0935085,15.093072
+1106,Grounding & Counting,objects365_qa,-2.2964675,17.525566
+1107,Grounding & Counting,objects365_qa,-3.1879425,13.193877
+1108,Grounding & Counting,objects365_qa,-2.6760135,13.829597
+1109,Grounding & Counting,objects365_qa,-2.1775434,12.870947
+1110,Grounding & Counting,objects365_qa,-1.1855267,11.783676
+1111,Grounding & Counting,objects365_qa,-1.1836683,16.344896
+1112,Grounding & Counting,objects365_qa,-1.608556,15.090458
+1113,Grounding & Counting,objects365_qa,-2.1136355,14.16562
+1114,Grounding & Counting,objects365_qa,-2.3335948,12.978815
+1115,Grounding & Counting,objects365_qa,-1.3887163,15.111507
+1116,Grounding & Counting,objects365_qa,0.94732374,13.842146
+1117,Grounding & Counting,objects365_qa,0.13381222,13.773756
+1118,Grounding & Counting,objects365_qa,-2.2739022,13.624816
+1119,Grounding & Counting,objects365_qa,-1.2355436,13.498342
+1120,Grounding & Counting,objects365_qa,-0.15974414,13.139577
+1121,Grounding & Counting,objects365_qa,-1.6127833,13.940305
+1122,Grounding & Counting,objects365_qa,-2.1489656,14.200482
+1123,Grounding & Counting,objects365_qa,-1.1124212,12.666072
+1124,Grounding & Counting,objects365_qa,-1.7189448,14.657562
+1125,Grounding & Counting,objects365_qa,-1.7964478,13.765957
+1126,Grounding & Counting,objects365_qa,-1.3854018,15.779286
+1127,Grounding & Counting,objects365_qa,-4.5416627,12.6137295
+1128,Grounding & Counting,objects365_qa,-0.5833148,12.687802
+1129,Grounding & Counting,objects365_qa,0.7002456,12.076195
+1130,Grounding & Counting,objects365_qa,-1.6218314,12.6297
+1131,Grounding & Counting,objects365_qa,-2.0090866,13.144378
+1132,Grounding & Counting,objects365_qa,-3.8905873,14.09144
+1133,Grounding & Counting,objects365_qa,-1.4862162,14.300795
+1134,Grounding & Counting,objects365_qa,-1.6721734,14.265931
+1135,Grounding & Counting,objects365_qa,-2.7549798,13.837875
+1136,Grounding & Counting,objects365_qa,-2.3494482,12.790322
+1137,Grounding & Counting,objects365_qa,-2.4970534,12.657688
+1138,Grounding & Counting,objects365_qa,-1.9969007,14.254142
+1139,Grounding & Counting,objects365_qa,-3.7182245,13.415132
+1140,Grounding & Counting,objects365_qa,-3.9991708,15.441768
+1141,Grounding & Counting,objects365_qa,-1.4131571,13.237267
+1142,Grounding & Counting,objects365_qa,-3.8476307,15.02824
+1143,Grounding & Counting,objects365_qa,-1.8533676,12.557681
+1144,Grounding & Counting,objects365_qa,-3.377942,14.089275
+1145,Grounding & Counting,objects365_qa,-2.9351132,14.839039
+1146,Grounding & Counting,objects365_qa,-3.2833838,14.310061
+1147,Grounding & Counting,objects365_qa,-3.7774377,14.291058
+1148,Grounding & Counting,objects365_qa,-0.9627761,13.561008
+1149,Grounding & Counting,objects365_qa,-1.8591784,14.181303
+1150,Grounding & Counting,objects365_qa,-4.2086587,13.715888
+1151,Grounding & Counting,objects365_qa,-2.688182,12.901205
+1152,Grounding & Counting,objects365_qa,-2.0203125,13.218085
+1153,Grounding & Counting,objects365_qa,-1.58034,14.475472
+1154,Grounding & Counting,objects365_qa,-1.2474319,14.134541
+1155,Grounding & Counting,objects365_qa,-1.8897669,13.623711
+1156,Grounding & Counting,objects365_qa,-2.1642325,13.257188
+1157,Grounding & Counting,objects365_qa,-1.4644091,15.231116
+1158,Grounding & Counting,objects365_qa,-5.7716146,14.677381
+1159,Grounding & Counting,objects365_qa,-2.2072744,13.503938
+1160,Grounding & Counting,objects365_qa,-2.851674,13.674607
+1161,Grounding & Counting,objects365_qa,-1.383384,13.076091
+1162,Grounding & Counting,objects365_qa,-1.9409691,15.656591
+1163,Grounding & Counting,objects365_qa,-3.976342,13.823318
+1164,Grounding & Counting,objects365_qa,-2.605507,15.237582
+1165,Grounding & Counting,objects365_qa,-1.4417936,13.25855
+1166,Grounding & Counting,objects365_qa,-5.226243,14.119703
+1167,Grounding & Counting,objects365_qa,-1.7077293,15.207848
+1168,Grounding & Counting,objects365_qa,-3.0913677,12.239021
+1169,Grounding & Counting,objects365_qa,-3.1178901,14.280295
+1170,Grounding & Counting,objects365_qa,-2.7813773,13.500068
+1171,Grounding & Counting,objects365_qa,-4.622278,12.665656
+1172,Grounding & Counting,objects365_qa,-1.9245214,15.175358
+1173,Grounding & Counting,objects365_qa,-1.8205452,13.3765
+1174,Grounding & Counting,objects365_qa,-1.1137084,14.370906
+1175,Grounding & Counting,objects365_qa,-4.2888174,13.73843
+1176,Grounding & Counting,objects365_qa,-1.0974977,16.10287
+1177,Grounding & Counting,objects365_qa,-0.9506335,13.492886
+1178,Grounding & Counting,objects365_qa,-1.8501304,14.297126
+1179,Grounding & Counting,objects365_qa,-4.523889,12.863494
+1180,Grounding & Counting,objects365_qa,-2.994191,13.264973
+1181,Grounding & Counting,objects365_qa,-0.74796754,13.956447
+1182,Grounding & Counting,objects365_qa,-0.47062624,14.041759
+1183,Grounding & Counting,objects365_qa,0.7174802,14.292761
+1184,Grounding & Counting,objects365_qa,-3.6827552,13.194036
+1185,Grounding & Counting,objects365_qa,-1.881917,14.973077
+1186,Grounding & Counting,objects365_qa,-4.3784404,14.532248
+1187,Grounding & Counting,objects365_qa,-4.0851846,13.943004
+1188,Grounding & Counting,objects365_qa,-1.9045763,13.20688
+1189,Grounding & Counting,objects365_qa,-0.9075715,13.416754
+1190,Grounding & Counting,objects365_qa,-0.93220145,12.436049
+1191,Grounding & Counting,objects365_qa,-2.6171033,15.490022
+1192,Grounding & Counting,objects365_qa,-0.4547748,14.817937
+1193,Grounding & Counting,objects365_qa,-1.1441321,10.734967
+1194,Grounding & Counting,objects365_qa,-1.7447535,14.138128
+1195,Grounding & Counting,objects365_qa,-2.3643475,13.932758
+1196,Grounding & Counting,objects365_qa,-2.0140176,13.519719
+1197,Grounding & Counting,objects365_qa,-1.2852308,14.746016
+1198,Grounding & Counting,objects365_qa,-2.0809596,12.535131
+1199,Grounding & Counting,objects365_qa,0.05504959,13.4964
+1200,Grounding & Counting,objects365_qa,-0.39000073,13.771782
+1201,Grounding & Counting,objects365_qa,-2.5437257,14.1807575
+1202,Grounding & Counting,objects365_qa,-0.6929019,14.870517
+1203,Grounding & Counting,objects365_qa,-3.8306763,14.742004
+1204,Grounding & Counting,objects365_qa,-3.7409532,14.8793335
+1205,Grounding & Counting,objects365_qa,-3.6435275,13.726134
+1206,Grounding & Counting,objects365_qa,-1.8782372,12.95533
+1207,Grounding & Counting,objects365_qa,-0.8780063,14.181423
+1208,Grounding & Counting,objects365_qa,-2.0596395,15.234883
+1209,Grounding & Counting,objects365_qa,-1.8798076,14.405055
+1210,Grounding & Counting,objects365_qa,-2.9574623,12.611294
+1211,Grounding & Counting,objects365_qa,-2.1968482,12.466001
+1212,Grounding & Counting,objects365_qa,-2.0306077,14.367065
+1213,Grounding & Counting,objects365_qa,-2.0694554,10.285254
+1214,Grounding & Counting,objects365_qa,-3.7937763,13.594311
+1215,Grounding & Counting,objects365_qa,-3.4082723,14.569007
+1216,Grounding & Counting,objects365_qa,-2.364398,14.279581
+1217,Grounding & Counting,objects365_qa,-1.0310595,13.788844
+1218,Grounding & Counting,objects365_qa,-6.052819,13.944904
+1219,Grounding & Counting,objects365_qa,-2.8022115,15.750512
+1220,Grounding & Counting,objects365_qa,-1.6662054,12.892409
+1221,Grounding & Counting,objects365_qa,-2.0090146,13.627944
+1222,Grounding & Counting,objects365_qa,-3.5679624,14.304952
+1223,Grounding & Counting,objects365_qa,-3.110716,13.4378195
+1224,Grounding & Counting,objects365_qa,-3.1788144,15.573506
+1225,Grounding & Counting,objects365_qa,-2.603618,13.105826
+1226,Grounding & Counting,objects365_qa,-1.6349196,12.953077
+1227,Grounding & Counting,objects365_qa,-1.485103,13.0733595
+1228,Grounding & Counting,objects365_qa,-1.7310972,14.280263
+1229,Grounding & Counting,objects365_qa,-3.0715816,12.784659
+1230,Grounding & Counting,objects365_qa,-2.9485266,14.75535
+1231,Grounding & Counting,objects365_qa,-3.751486,15.063919
+1232,Grounding & Counting,objects365_qa,-3.4424183,13.815565
+1233,Grounding & Counting,objects365_qa,-1.9569427,13.593262
+1234,Grounding & Counting,objects365_qa,-2.4116402,14.1085825
+1235,Grounding & Counting,objects365_qa,0.12328538,13.031088
+1236,Grounding & Counting,objects365_qa,-1.8039991,12.859735
+1237,Grounding & Counting,objects365_qa,-1.0692085,14.436484
+1238,Grounding & Counting,objects365_qa,-1.1507427,14.4154825
+1239,Grounding & Counting,objects365_qa,-2.1137578,11.6871
+1240,Grounding & Counting,objects365_qa,-2.55026,12.235803
+1241,Grounding & Counting,objects365_qa,-2.284663,13.745578
+1242,Grounding & Counting,objects365_qa,-3.65729,13.914726
+1243,Grounding & Counting,objects365_qa,-1.3133111,12.592305
+1244,Grounding & Counting,objects365_qa,-2.03134,12.572514
+1245,Grounding & Counting,objects365_qa,-1.2027699,14.195614
+1246,Grounding & Counting,objects365_qa,-1.4905783,14.204572
+1247,Grounding & Counting,objects365_qa,-1.4291193,14.005105
+1248,Grounding & Counting,objects365_qa,-0.18903129,14.6363125
+1249,OCR QA,ocrvqa,-3.007456,15.424597
+1250,OCR QA,ocrvqa,-1.9252207,15.831923
+1251,OCR QA,ocrvqa,-2.7130177,14.887906
+1252,OCR QA,ocrvqa,-3.404401,16.228456
+1253,OCR QA,ocrvqa,-1.4823917,15.27851
+1254,OCR QA,ocrvqa,-3.2153766,15.188947
+1255,OCR QA,ocrvqa,-3.1798456,15.848485
+1256,OCR QA,ocrvqa,-4.0313406,14.592979
+1257,OCR QA,ocrvqa,-4.8766723,17.622332
+1258,OCR QA,ocrvqa,-2.6125069,14.824715
+1259,OCR QA,ocrvqa,-2.342191,15.426579
+1260,OCR QA,ocrvqa,-2.290442,14.894173
+1261,OCR QA,ocrvqa,-2.9414363,15.028527
+1262,OCR QA,ocrvqa,-1.4030048,15.192581
+1263,OCR QA,ocrvqa,-1.8940259,15.5841055
+1264,OCR QA,ocrvqa,-2.1474206,14.668903
+1265,Naive OCR,olmOCR-mix-0225-books,-0.049820628,18.168592
+1266,Naive OCR,olmOCR-mix-0225-documents,-1.3186306,18.931374
+1267,Naive OCR,olmOCR-mix-0225-documents,-1.5079118,19.591742
+1268,Naive OCR,olmOCR-mix-0225-documents,-1.9480561,17.993065
+1269,Naive OCR,olmOCR-mix-0225-documents,-1.5744836,19.130222
+1270,Naive OCR,olmOCR-mix-0225-documents,-1.8590188,19.165867
+1271,Naive OCR,olmOCR-mix-0225-documents,-0.2036161,17.015211
+1272,Naive OCR,olmOCR-mix-0225-documents,-1.8654215,19.72361
+1273,Naive OCR,olmOCR-mix-0225-documents,-3.48016,17.774706
+1274,Naive OCR,olmOCR-mix-0225-documents,-1.4446557,17.443485
+1275,Naive OCR,olmOCR-mix-0225-documents,-1.3327416,17.503735
+1276,Naive OCR,olmOCR-mix-0225-documents,-0.9926907,18.439743
+1277,Naive OCR,olmOCR-mix-0225-documents,-1.7444654,17.647224
+1278,Naive OCR,olmOCR-mix-0225-documents,-3.6816046,18.750822
+1279,Naive OCR,olmOCR-mix-0225-documents,-3.4683683,18.11246
+1280,Naive OCR,olmOCR-mix-0225-documents,-2.4299297,19.77487
+1281,Naive OCR,olmOCR-mix-0225-documents,-1.5804191,19.438263
+1282,Naive OCR,olmOCR-mix-0225-documents,-2.2610602,17.83327
+1283,Naive OCR,olmOCR-mix-0225-documents,-3.0591917,19.259407
+1284,Naive OCR,olmOCR-mix-0225-documents,-1.7917006,17.772846
+1285,Naive OCR,olmOCR-mix-0225-documents,-1.7828513,18.87353
+1286,Naive OCR,olmOCR-mix-0225-documents,-1.2982534,19.311142
+1287,Naive OCR,olmOCR-mix-0225-documents,-1.481188,18.249447
+1288,Grounding & Counting,oodvqa,-0.6815327,13.228337
+1289,Naive OCR,orand_car_a,0.8555107,12.444546
+1290,Science,pathvqa,-2.7144184,13.87168
+1291,Science,pathvqa,0.960431,12.806884
+1292,Science,pathvqa,-3.0353422,15.271672
+1293,OCR QA,pdfvqa,-1.656308,19.853363
+1294,Chart & Table,plotqa,-7.0799775,14.756452
+1295,Chart & Table,plotqa,-5.5909977,12.975237
+1296,Chart & Table,plotqa,-5.918642,13.3526325
+1297,Chart & Table,plotqa,-6.8707027,14.74776
+1298,Chart & Table,plotqa,-6.968584,13.844693
+1299,Chart & Table,plotqa,-7.1763444,14.717227
+1300,Chart & Table,plotqa,-7.027178,14.43275
+1301,Chart & Table,plotqa,-5.5268917,13.130444
+1302,Chart & Table,plotqa,-7.063167,14.424732
+1303,Chart & Table,plotqa,-6.9672456,14.123852
+1304,Chart & Table,plotqa,-7.039555,14.098095
+1305,Chart & Table,plotqa,-7.0466204,14.118572
+1306,Chart & Table,plotqa,-7.199804,14.048889
+1307,Chart & Table,plotqa,-7.206863,13.914144
+1308,Chart & Table,plotqa,-7.2543044,14.671591
+1309,Science,pmc_vqa(mathv360k),-1.5792531,15.700976
+1310,Science,pmc_vqa(mathv360k),-3.2694328,14.33561
+1311,Science,pmc_vqa(mathv360k),-1.3354739,15.164038
+1312,Mathematics,raven,-4.67516,13.810964
+1313,Mathematics,raven,-4.82738,14.449474
+1314,Mathematics,raven,-4.6267257,13.246527
+1315,Mathematics,raven,-5.0504494,14.91181
+1316,Naive OCR,rendered_text,-3.238513,10.913463
+1317,Chart & Table,robut_sqa,-4.80446,16.619642
+1318,Chart & Table,robut_wikisql,-4.9652443,16.691702
+1319,Chart & Table,robut_wikisql,-4.864325,16.472876
+1320,Chart & Table,robut_wikisql,-4.7404175,16.573008
+1321,Chart & Table,robut_wikisql,-5.0994334,16.797874
+1322,Chart & Table,robut_wikisql,-4.9155374,16.536234
+1323,Chart & Table,robut_wikisql,-4.634686,16.658693
+1324,Chart & Table,robut_wikisql,-4.7533784,16.742083
+1325,Chart & Table,robut_wtq,-4.98517,16.481878
+1326,Chart & Table,robut_wtq,-4.6912847,16.624222
+1327,Chart & Table,robut_wtq,-4.755297,16.722153
+1328,Science,scienceqa,-5.0773787,15.692775
+1329,Science,scienceqa(nona_context),-4.7690954,15.214425
+1330,OCR QA,screen2words,-3.631612,16.746061
+1331,OCR QA,screenqa,-3.4821894,17.005753
+1332,OCR QA,screenqa,-3.8667197,17.089247
+1333,OCR QA,screenqa,-3.9886274,16.325684
+1334,OCR QA,screenqa,-3.6292388,16.64084
+1335,OCR QA,screenqa,-3.867926,16.939747
+1336,OCR QA,screenqa,-4.2676463,16.212013
+1337,OCR QA,screenqa,-2.2830477,16.461208
+1338,OCR QA,screenqa,-2.7972887,16.759937
+1339,Captioning & Knowledge,sharegpt4o,-1.028643,14.316716
+1340,Captioning & Knowledge,sharegpt4o,-0.63227594,18.262432
+1341,Captioning & Knowledge,sharegpt4o,-5.5509615,13.385918
+1342,Captioning & Knowledge,sharegpt4o,-2.9258623,12.430899
+1343,Captioning & Knowledge,sharegpt4o,-1.752485,9.818922
+1344,Captioning & Knowledge,sharegpt4v(coco),-4.396255,15.497202
+1345,Captioning & Knowledge,sharegpt4v(coco),-6.7254567,13.543941
+1346,Captioning & Knowledge,sharegpt4v(coco),-3.4319088,13.363286
+1347,Captioning & Knowledge,sharegpt4v(coco),-0.94431686,17.7518
+1348,Captioning & Knowledge,sharegpt4v(coco),-3.9650493,14.540166
+1349,Captioning & Knowledge,sharegpt4v(knowledge),-2.752691,12.089081
+1350,Captioning & Knowledge,sharegpt4v(llava),-2.5029697,12.902743
+1351,Captioning & Knowledge,sharegpt4v(llava),1.3518825,13.444673
+1352,Captioning & Knowledge,sharegpt4v(sam),-3.3568525,12.670897
+1353,General VQA,sketchyvqa,-4.8872657,11.832038
+1354,OCR QA,slidevqa,-3.9107704,15.906221
+1355,General VQA,spark,-4.3768535,14.491207
+1356,General VQA,spatialsense,-3.263734,14.976058
+1357,General VQA,spot_the_diff,-1.5317576,16.37698
+1358,Naive OCR,sroie,1.3219887,11.1940775
+1359,Naive OCR,sroie,1.5913999,11.914926
+1360,Naive OCR,sroie,1.9426727,11.218723
+1361,OCR QA,st_vqa,-2.9044545,15.1445875
+1362,OCR QA,sujet_finance,-1.5862796,17.535973
+1363,Mathematics,super_clevr(mathv360k),-1.9466949,10.134516
+1364,Naive OCR,svrd,-0.71856314,17.167393
+1365,Naive OCR,synthdog,-3.502172,10.745516
+1366,Naive OCR,synthdog,-3.0102074,10.791281
+1367,Naive OCR,synthdog,-3.6376667,10.726992
+1368,Naive OCR,synthdog,-2.9797876,11.0659275
+1369,Naive OCR,synthdog,-3.2355602,11.539314
+1370,Naive OCR,synthdog,-2.9555662,11.265699
+1371,Naive OCR,synthdog,-2.96735,10.951926
+1372,Naive OCR,synthdog,-2.6145597,11.390062
+1373,Naive OCR,synthdog,-2.5206306,11.208609
+1374,Naive OCR,synthdog,-3.2974412,10.964338
+1375,Naive OCR,synthdog,-3.0609767,10.871874
+1376,Naive OCR,synthdog,-3.7733376,10.823489
+1377,Naive OCR,synthdog,-3.5317874,10.925116
+1378,Naive OCR,synthdog,-3.5101364,10.958425
+1379,Naive OCR,synthdog,-3.8239758,10.659712
+1380,Naive OCR,synthdog,-3.0691514,11.058068
+1381,Naive OCR,synthdog,-2.753329,11.22186
+1382,Naive OCR,synthdog,-2.6324878,11.007256
+1383,Naive OCR,synthdog,-2.6753669,10.980343
+1384,Naive OCR,synthdog,-3.1924512,11.233193
+1385,Naive OCR,synthdog,-2.7523263,11.363193
+1386,Naive OCR,synthdog,-3.5240006,10.827522
+1387,Naive OCR,synthdog,-2.8060973,10.9742775
+1388,Naive OCR,synthdog,-3.7526023,10.584175
+1389,Naive OCR,synthdog,-2.8121305,10.815339
+1390,Naive OCR,synthdog,-3.439075,10.882955
+1391,Naive OCR,synthdog,-3.2205725,10.566704
+1392,Naive OCR,synthdog,-3.7893758,11.147975
+1393,Naive OCR,synthdog,-3.536549,11.21009
+1394,Naive OCR,synthdog,-3.1966834,11.723928
+1395,Naive OCR,synthdog,-2.6977828,11.074001
+1396,Naive OCR,synthdog,-3.2249844,11.257024
+1397,Naive OCR,synthdog,-2.8481858,11.527405
+1398,Naive OCR,synthdog,-3.1936653,11.021578
+1399,Naive OCR,synthdog,-3.1159499,10.622795
+1400,Naive OCR,synthdog,-3.3986914,11.110819
+1401,Naive OCR,synthdog,-3.3330255,10.823691
+1402,Naive OCR,synthdog,-3.4130352,10.685922
+1403,Naive OCR,synthdog,-3.448721,11.279447
+1404,Naive OCR,synthdog,-3.3531446,10.8329
+1405,Naive OCR,synthdog,-2.4609246,11.361593
+1406,Naive OCR,synthdog,-3.1755846,10.649888
+1407,Naive OCR,synthdog,-3.0586236,11.307644
+1408,Naive OCR,synthdog,-3.1265254,10.875503
+1409,Naive OCR,synthdog,-3.6899135,10.746278
+1410,Naive OCR,synthdog,-3.3908458,11.223358
+1411,Naive OCR,synthdog,-3.3751297,10.516364
+1412,Naive OCR,synthdog,-3.488507,10.531199
+1413,Naive OCR,synthdog,-2.8473232,11.15339
+1414,Naive OCR,synthdog,-3.20954,11.064686
+1415,Chart & Table,tabmwp,-0.58479166,15.880317
+1416,Chart & Table,tabmwp,-0.2904701,15.783937
+1417,Chart & Table,tabmwp(mathv360k),-0.7231676,15.953247
+1418,Chart & Table,tabmwp(mathv360k),-0.2565399,15.977477
+1419,Naive OCR,tal_ocr_eng,1.5386844,10.62674
+1420,Naive OCR,tal_ocr_eng,1.6549606,11.601942
+1421,Naive OCR,tal_ocr_eng,1.4002947,11.764387
+1422,Naive OCR,tal_ocr_eng,1.2745994,11.2013035
+1423,Naive OCR,tal_ocr_eng,1.5937434,10.996404
+1424,Naive OCR,tal_ocr_eng,2.2340503,11.136835
+1425,Naive OCR,tal_ocr_eng,2.127229,11.4193945
+1426,Naive OCR,tal_ocr_eng,2.2122283,11.736268
+1427,Naive OCR,tal_ocr_eng,1.9461217,11.907448
+1428,Naive OCR,tal_ocr_eng,1.309156,12.007619
+1429,Naive OCR,tal_ocr_eng,2.2799487,11.581288
+1430,Naive OCR,tal_ocr_eng,2.1628947,10.770825
+1431,Naive OCR,tal_ocr_eng,2.1115048,11.302841
+1432,Naive OCR,tal_ocr_eng,1.8126061,11.708756
+1433,Naive OCR,tal_ocr_eng,2.1145957,10.837662
+1434,Naive OCR,tal_ocr_eng,1.3191409,11.1685295
+1435,Naive OCR,tal_ocr_eng,1.9186009,10.96969
+1436,Naive OCR,tal_ocr_eng,1.1728864,11.00445
+1437,Naive OCR,tal_ocr_eng,1.2456665,11.044403
+1438,Naive OCR,tal_ocr_eng,2.0839403,10.841493
+1439,Naive OCR,tal_ocr_eng,1.932624,11.314103
+1440,Naive OCR,tal_ocr_eng,2.029552,10.737507
+1441,Naive OCR,tal_ocr_eng,2.1489453,11.5904
+1442,Naive OCR,tal_ocr_eng,2.1690724,12.882016
+1443,Naive OCR,tal_ocr_eng,1.3835964,12.141184
+1444,Grounding & Counting,tallyqa,-2.6993341,14.153768
+1445,Grounding & Counting,tallyqa,-2.9867494,13.562643
+1446,Grounding & Counting,tallyqa,-1.7105672,13.957441
+1447,Grounding & Counting,tallyqa,-2.2648647,13.452209
+1448,Grounding & Counting,tallyqa,-0.5068742,12.945275
+1449,Grounding & Counting,tallyqa,-1.4287742,15.124818
+1450,Grounding & Counting,tallyqa,-1.191937,12.840298
+1451,Grounding & Counting,tallyqa,-0.7951459,13.128243
+1452,Grounding & Counting,tallyqa,-2.041517,14.90799
+1453,Chart & Table,tat_dqa,-2.8914964,18.513271
+1454,Chart & Table,tat_qa,-4.983957,16.240915
+1455,Captioning & Knowledge,textcaps,-0.7445523,15.09082
+1456,Captioning & Knowledge,textcaps,-1.1821718,13.070472
+1457,OCR QA,textocr(gpt4v),-1.6652915,13.593408
+1458,OCR QA,textocr(gpt4v),-1.2120732,14.200418
+1459,OCR QA,textvqa,-3.954238,15.533072
+1460,OCR QA,textvqa,-3.5793169,12.164482
+1461,Science,tqa,-3.1680925,12.166714
+1462,Mathematics,unigeo(mathv360k),-4.861552,11.670065
+1463,OCR QA,ureader_cap,-3.1146836,16.865677
+1464,OCR QA,ureader_cap,-1.4596508,14.073721
+1465,OCR QA,ureader_cap,-1.066469,12.193185
+1466,OCR QA,ureader_cap,-2.4253721,16.439564
+1467,OCR QA,ureader_cap,-1.4031659,13.693128
+1468,OCR QA,ureader_cap,-2.7545621,15.230273
+1469,OCR QA,ureader_cap,-3.8215294,16.937256
+1470,OCR QA,ureader_cap,-3.0048773,12.195416
+1471,OCR QA,ureader_cap,-2.9467802,14.161315
+1472,OCR QA,ureader_ie,-3.3650353,18.916605
+1473,OCR QA,ureader_kg_processed,-2.6326602,17.692429
+1474,OCR QA,ureader_kg_processed,-3.7576883,16.363173
+1475,OCR QA,ureader_kg_processed,-7.405658,15.387937
+1476,OCR QA,ureader_qa_processed,-0.38821623,16.618586
+1477,OCR QA,ureader_qa_processed,0.34040427,16.536598
+1478,OCR QA,ureader_qa_processed,-1.2258626,15.028207
+1479,OCR QA,ureader_qa_processed,0.12460345,16.11192
+1480,OCR QA,ureader_qa_processed,0.42298225,16.552795
+1481,OCR QA,ureader_qa_processed,-2.852388,16.006634
+1482,OCR QA,ureader_qa_processed,-0.6593659,13.979379
+1483,OCR QA,ureader_qa_processed,-2.9657156,18.554277
+1484,OCR QA,ureader_qa_processed,0.15679286,16.374321
+1485,OCR QA,ureader_qa_processed,0.0874012,16.609613
+1486,OCR QA,ureader_qa_processed,-7.2615995,14.888373
+1487,OCR QA,ureader_qa_processed,-0.548176,18.47858
+1488,OCR QA,ureader_qa_processed,-7.314555,15.370032
+1489,OCR QA,ureader_qa_processed,0.042276993,16.33529
+1490,OCR QA,ureader_qa_processed,0.458077,16.473448
+1491,OCR QA,ureader_qa_processed,0.20129277,16.408411
+1492,OCR QA,ureader_qa_processed,-1.3313242,16.77525
+1493,OCR QA,ureader_qa_processed,-1.3548806,17.374992
+1494,OCR QA,ureader_qa_processed,0.28216,16.484983
+1495,OCR QA,ureader_qa_processed,0.31041244,16.489374
+1496,OCR QA,ureader_qa_processed,0.40330794,16.222654
+1497,OCR QA,ureader_qa_processed,-1.5861939,13.4266615
+1498,OCR QA,ureader_qa_processed,-1.4182721,18.93737
+1499,OCR QA,ureader_qa_processed,-2.6782496,18.193298
+1500,OCR QA,ureader_qa_processed,-0.87527555,16.933899
+1501,General VQA,vision_flan(filtered),-0.56303865,13.4160385
+1502,General VQA,vision_flan(filtered),-5.3517127,17.694803
+1503,General VQA,vision_flan(filtered),-3.2433019,15.603335
+1504,General VQA,vision_flan(filtered),-1.8283677,9.439475
+1505,General VQA,vision_flan(filtered),-3.5837595,11.817849
+1506,General VQA,vision_flan(filtered),-1.9949989,9.659579
+1507,General VQA,vision_flan(filtered),-3.9365442,15.855789
+1508,General VQA,vision_flan(filtered),-3.4549484,17.179455
+1509,General VQA,vision_flan(filtered),-4.505904,15.089286
+1510,General VQA,vision_flan(filtered),-2.2987983,12.389837
+1511,General VQA,vision_flan(filtered),-1.1805819,15.897051
+1512,General VQA,vision_flan(filtered),-0.7427064,12.143822
+1513,General VQA,vision_flan(filtered),-1.5613407,14.766336
+1514,General VQA,vision_flan(filtered),-5.431604,16.530396
+1515,General VQA,vision_flan(filtered),-2.1079543,9.795982
+1516,General VQA,vision_flan(filtered),-4.0272603,12.842213
+1517,General VQA,vision_flan(filtered),-0.7639318,14.016364
+1518,Chart & Table,vistext,-4.615444,15.827518
+1519,General VQA,visual7w,-4.9198766,12.580895
+1520,OCR QA,visualmrc,-3.1618717,17.868393
+1521,Science,visualwebinstruct(filtered),0.19766942,16.277098
+1522,Science,visualwebinstruct(filtered),-5.207955,12.296344
+1523,Science,visualwebinstruct(filtered),-2.478752,15.9534235
+1524,Science,visualwebinstruct(filtered),-5.3273616,13.613769
+1525,Science,visualwebinstruct(filtered),-5.063052,12.780937
+1526,Science,visualwebinstruct(filtered),-4.5611515,11.5869
+1527,Science,visualwebinstruct(filtered),-6.2776504,12.762167
+1528,Science,visualwebinstruct(filtered),-5.0601234,15.37361
+1529,Science,visualwebinstruct(filtered),-4.8031926,13.887086
+1530,Science,visualwebinstruct(filtered),-2.041472,15.937402
+1531,Science,visualwebinstruct(filtered),-5.0025134,12.591787
+1532,Science,visualwebinstruct(filtered),-4.588724,11.807936
+1533,Science,visualwebinstruct(filtered),-3.2328036,15.801565
+1534,Science,visualwebinstruct(filtered),-4.493823,12.591351
+1535,Science,visualwebinstruct(filtered),-4.500083,11.668415
+1536,Science,visualwebinstruct(filtered),-5.212913,13.715419
+1537,Science,visualwebinstruct(filtered),-4.4452915,12.369002
+1538,Science,visualwebinstruct(filtered),-5.3771167,12.082145
+1539,Science,visualwebinstruct(filtered),-3.0751553,15.810075
+1540,Science,visualwebinstruct(filtered),-5.6383653,11.890171
+1541,Science,visualwebinstruct(filtered),0.16722926,16.123463
+1542,Science,visualwebinstruct(filtered),-5.635973,11.784142
+1543,Science,visualwebinstruct(filtered),-4.7131815,11.859157
+1544,Science,visualwebinstruct(filtered),-6.674552,16.054293
+1545,Science,visualwebinstruct(filtered),0.31570235,16.438677
+1546,Science,visualwebinstruct(filtered),-4.5665827,12.13578
+1547,General VQA,vizwiz(mathv360k),-1.414545,13.934932
+1548,Chart & Table,vqaonbd,-0.8374918,16.678001
+1549,Chart & Table,vqaonbd,-0.76935184,16.891268
+1550,Chart & Table,vqaonbd,-0.7749725,16.949081
+1551,Science,vqarad,-3.3113384,15.209322
+1552,General VQA,vqav2,-0.81680644,13.298286
+1553,General VQA,vqav2,-3.7252114,14.326692
+1554,General VQA,vqav2,-3.561089,13.907111
+1555,General VQA,vqav2,-3.2257302,12.510049
+1556,General VQA,vqav2,-2.9566333,12.515226
+1557,General VQA,vqav2,-1.6344411,13.967026
+1558,General VQA,vqav2,-1.4874606,13.836463
+1559,General VQA,vqav2,-4.3959174,12.975297
+1560,General VQA,vsr,-1.7102809,14.585667
+1561,General VQA,websight,-3.1001532,18.220957
+1562,General VQA,wildvision,-1.8356924,14.7251425
+1563,Naive OCR,wordart,-6.515975,13.7404175
+1564,General VQA,yesbut,1.4558206,10.848951

data/data.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6bbaf02f1b470da41754e3828e81e76ef386d9b3cfb8b57dcc7cbfd4225956cc
+size 14121778

data/font-sprite-mapping.json

@@ -0,0 +1,295 @@
+{
+  "abeezee": "abeezee_a",
+  "abel": "abel_a",
+  "abhaya libre": "abhaya_libre_a",
+  "aboreto": "aboreto_a",
+  "abril fatface": "abril_fatface_a",
+  "abyssinica sil": "abyssinica_sil_a",
+  "aclonica": "aclonica_a",
+  "acme": "acme_a",
+  "actor": "actor_a",
+  "adamina": "adamina_a",
+  "adlam display": "adlam_display_a",
+  "advent pro": "advent_pro_a",
+  "afacad": "afacad_a",
+  "afacad flux": "afacad_flux_a",
+  "agbalumo": "agbalumo_a",
+  "agdasima": "agdasima_a",
+  "agu display": "agu_display_a",
+  "aguafina script": "aguafina_script_a",
+  "akatab": "akatab_a",
+  "akaya kanadaka": "akaya_kanadaka_a",
+  "akaya telivigala": "akaya_telivigala_a",
+  "akronim": "akronim_a",
+  "akshar": "akshar_a",
+  "aladin": "aladin_a",
+  "alata": "alata_a",
+  "alatsi": "alatsi_a",
+  "albert sans": "albert_sans_a",
+  "aldrich": "aldrich_a",
+  "alef": "alef_a",
+  "alegreya": "alegreya_a",
+  "alegreya sans": "alegreya_sans_a",
+  "alegreya sans sc": "alegreya_sans_sc_a",
+  "alegreya sc": "alegreya_sc_a",
+  "aleo": "aleo_a",
+  "alex brush": "alex_brush_a",
+  "alexandria": "alexandria_a",
+  "alfa slab one": "alfa_slab_one_a",
+  "alice": "alice_a",
+  "alike": "alike_a",
+  "alike angular": "alike_angular_a",
+  "alkalami": "alkalami_a",
+  "alkatra": "alkatra_a",
+  "allan": "allan_a",
+  "allerta": "allerta_a",
+  "allerta stencil": "allerta_stencil_a",
+  "allison": "allison_a",
+  "allura": "allura_a",
+  "almarai": "almarai_a",
+  "almendra": "almendra_a",
+  "almendra display": "almendra_display_a",
+  "almendra sc": "almendra_sc_a",
+  "alumni sans": "alumni_sans_a",
+  "alumni sans collegiate one": "alumni_sans_collegiate_one_a",
+  "alumni sans pinstripe": "alumni_sans_pinstripe_a",
+  "alumni sans sc": "alumni_sans_sc_a",
+  "amarante": "amarante_a",
+  "amaranth": "amaranth_a",
+  "amatic sc": "amatic_sc_a",
+  "amethysta": "amethysta_a",
+  "amiko": "amiko_a",
+  "amiri": "amiri_a",
+  "amiri quran": "amiri_quran_a",
+  "amita": "amita_a",
+  "anaheim": "anaheim_a",
+  "ancizar sans": "ancizar_sans_a",
+  "ancizar serif": "ancizar_serif_a",
+  "andada pro": "andada_pro_a",
+  "andika": "andika_a",
+  "anek bangla": "anek_bangla_a",
+  "anek devanagari": "anek_devanagari_a",
+  "anek gujarati": "anek_gujarati_a",
+  "anek gurmukhi": "anek_gurmukhi_a",
+  "anek kannada": "anek_kannada_a",
+  "anek latin": "anek_latin_a",
+  "anek malayalam": "anek_malayalam_a",
+  "anek odia": "anek_odia_a",
+  "anek tamil": "anek_tamil_a",
+  "anek telugu": "anek_telugu_a",
+  "angkor": "angkor_a",
+  "annapurna sil": "annapurna_sil_a",
+  "annie use your telescope": "annie_use_your_telescope_a",
+  "anonymous pro": "anonymous_pro_a",
+  "anta": "anta_a",
+  "antic": "antic_a",
+  "antic didone": "antic_didone_a",
+  "antic slab": "antic_slab_a",
+  "anton": "anton_a",
+  "anton sc": "anton_sc_a",
+  "antonio": "antonio_a",
+  "anuphan": "anuphan_a",
+  "anybody": "anybody_a",
+  "ar one sans": "ar_one_sans_a",
+  "arapey": "arapey_a",
+  "arbutus": "arbutus_a",
+  "arbutus slab": "arbutus_slab_a",
+  "architects daughter": "architects_daughter_a",
+  "archivo": "archivo_a",
+  "archivo black": "archivo_black_a",
+  "archivo narrow": "archivo_narrow_a",
+  "are you serious": "are_you_serious_a",
+  "aref ruqaa": "aref_ruqaa_a",
+  "aref ruqaa ink": "aref_ruqaa_ink_a",
+  "arima": "arima_a",
+  "arimo": "arimo_a",
+  "arizonia": "arizonia_a",
+  "armata": "armata_a",
+  "arsenal": "arsenal_a",
+  "arsenal sc": "arsenal_sc_a",
+  "artifika": "artifika_a",
+  "arvo": "arvo_a",
+  "arya": "arya_a",
+  "asap": "asap_a",
+  "asap condensed": "asap_condensed_a",
+  "asar": "asar_a",
+  "asimovian": "asimovian_a",
+  "asset": "asset_a",
+  "assistant": "assistant_a",
+  "asta sans": "asta_sans_a",
+  "astloch": "astloch_a",
+  "asul": "asul_a",
+  "athiti": "athiti_a",
+  "atkinson hyperlegible": "atkinson_hyperlegible_a",
+  "atkinson hyperlegible mono": "atkinson_hyperlegible_mono_a",
+  "atkinson hyperlegible next": "atkinson_hyperlegible_next_a",
+  "atma": "atma_a",
+  "atomic age": "atomic_age_a",
+  "aubrey": "aubrey_a",
+  "audiowide": "audiowide_a",
+  "autour one": "autour_one_a",
+  "average": "average_a",
+  "average sans": "average_sans_a",
+  "averia gruesa libre": "averia_gruesa_libre_a",
+  "averia libre": "averia_libre_a",
+  "averia sans libre": "averia_sans_libre_a",
+  "averia serif libre": "averia_serif_libre_a",
+  "azeret mono": "azeret_mono_a",
+  "b612": "b612_a",
+  "b612 mono": "b612_mono_a",
+  "babylonica": "babylonica_a",
+  "bacasime antique": "bacasime_antique_a",
+  "bad script": "bad_script_a",
+  "badeen display": "badeen_display_a",
+  "bagel fat one": "bagel_fat_one_a",
+  "bahiana": "bahiana_a",
+  "bahianita": "bahianita_a",
+  "bai jamjuree": "bai_jamjuree_a",
+  "bakbak one": "bakbak_one_a",
+  "ballet": "ballet_a",
+  "baloo 2": "baloo_2_a",
+  "baloo bhai 2": "baloo_bhai_2_a",
+  "baloo bhaijaan 2": "baloo_bhaijaan_2_a",
+  "baloo bhaina 2": "baloo_bhaina_2_a",
+  "baloo da 2": "baloo_da_2_a",
+  "baloo paaji 2": "baloo_paaji_2_a",
+  "baloo tamma 2": "baloo_tamma_2_a",
+  "baloo tammudu 2": "baloo_tammudu_2_a",
+  "baloo thambi 2": "baloo_thambi_2_a",
+  "balsamiq sans": "balsamiq_sans_a",
+  "bangers": "bangers_a",
+  "barlow": "barlow_a",
+  "barlow condensed": "barlow_condensed_a",
+  "barlow semi condensed": "barlow_semi_condensed_a",
+  "barriecito": "barriecito_a",
+  "barrio": "barrio_a",
+  "basic": "basic_a",
+  "baskervville": "baskervville_a",
+  "baskervville sc": "baskervville_sc_a",
+  "battambang": "battambang_a",
+  "baumans": "baumans_a",
+  "bayon": "bayon_a",
+  "be vietnam pro": "be_vietnam_pro_a",
+  "beau rivage": "beau_rivage_a",
+  "bebas neue": "bebas_neue_a",
+  "beiruti": "beiruti_a",
+  "belanosima": "belanosima_a",
+  "belgrano": "belgrano_a",
+  "bellefair": "bellefair_a",
+  "belleza": "belleza_a",
+  "bellota": "bellota_a",
+  "bellota text": "bellota_text_a",
+  "benchnine": "benchnine_a",
+  "benne": "benne_a",
+  "bentham": "bentham_a",
+  "berkshire swash": "berkshire_swash_a",
+  "besley": "besley_a",
+  "beth ellen": "beth_ellen_a",
+  "bevan": "bevan_a",
+  "bhutuka expanded one": "bhutuka_expanded_one_a",
+  "big shoulders": "big_shoulders_a",
+  "big shoulders inline": "big_shoulders_inline_a",
+  "big shoulders stencil": "big_shoulders_stencil_a",
+  "bigelow rules": "bigelow_rules_a",
+  "bigshot one": "bigshot_one_a",
+  "bilbo": "bilbo_a",
+  "bilbo swash caps": "bilbo_swash_caps_a",
+  "biorhyme": "biorhyme_a",
+  "biorhyme expanded": "biorhyme_expanded_a",
+  "birthstone": "birthstone_a",
+  "birthstone bounce": "birthstone_bounce_a",
+  "biryani": "biryani_a",
+  "bitcount": "bitcount_a",
+  "bitcount grid double": "bitcount_grid_double_a",
+  "bitcount grid double ink": "bitcount_grid_double_ink_a",
+  "bitcount grid single": "bitcount_grid_single_a",
+  "bitcount grid single ink": "bitcount_grid_single_ink_a",
+  "bitcount ink": "bitcount_ink_a",
+  "bitcount prop double": "bitcount_prop_double_a",
+  "bitcount prop double ink": "bitcount_prop_double_ink_a",
+  "bitcount prop single": "bitcount_prop_single_a",
+  "bitcount prop single ink": "bitcount_prop_single_ink_a",
+  "bitcount single": "bitcount_single_a",
+  "bitcount single ink": "bitcount_single_ink_a",
+  "bitter": "bitter_a",
+  "biz udgothic": "biz_udgothic_a",
+  "biz udmincho": "biz_udmincho_a",
+  "biz udpgothic": "biz_udpgothic_a",
+  "biz udpmincho": "biz_udpmincho_a",
+  "black and white picture": "black_and_white_picture_a",
+  "black han sans": "black_han_sans_a",
+  "black ops one": "black_ops_one_a",
+  "blaka": "blaka_a",
+  "blaka hollow": "blaka_hollow_a",
+  "blaka ink": "blaka_ink_a",
+  "blinker": "blinker_a",
+  "bodoni moda": "bodoni_moda_a",
+  "bodoni moda sc": "bodoni_moda_sc_a",
+  "bokor": "bokor_a",
+  "boldonse": "boldonse_a",
+  "bona nova": "bona_nova_a",
+  "bona nova sc": "bona_nova_sc_a",
+  "bonbon": "bonbon_a",
+  "bonheur royale": "bonheur_royale_a",
+  "boogaloo": "boogaloo_a",
+  "borel": "borel_a",
+  "bowlby one": "bowlby_one_a",
+  "bowlby one sc": "bowlby_one_sc_a",
+  "braah one": "braah_one_a",
+  "brawler": "brawler_a",
+  "bree serif": "bree_serif_a",
+  "bricolage grotesque": "bricolage_grotesque_a",
+  "bruno ace": "bruno_ace_a",
+  "bruno ace sc": "bruno_ace_sc_a",
+  "brygada 1918": "brygada_1918_a",
+  "bubblegum sans": "bubblegum_sans_a",
+  "bubbler one": "bubbler_one_a",
+  "buenard": "buenard_a",
+  "bungee": "bungee_a",
+  "bungee hairline": "bungee_hairline_a",
+  "bungee inline": "bungee_inline_a",
+  "bungee outline": "bungee_outline_a",
+  "bungee shade": "bungee_shade_a",
+  "bungee spice": "bungee_spice_a",
+  "bungee tint": "bungee_tint_a",
+  "butcherman": "butcherman_a",
+  "butterfly kids": "butterfly_kids_a",
+  "bytesized": "bytesized_a",
+  "cabin": "cabin_a",
+  "cabin condensed": "cabin_condensed_a",
+  "cabin sketch": "cabin_sketch_a",
+  "cactus classical serif": "cactus_classical_serif_a",
+  "caesar dressing": "caesar_dressing_a",
+  "cagliostro": "cagliostro_a",
+  "cairo": "cairo_a",
+  "cairo play": "cairo_play_a",
+  "cal sans": "cal_sans_a",
+  "caladea": "caladea_a",
+  "calligraffitti": "calligraffitti_a",
+  "cambay": "cambay_a",
+  "cambo": "cambo_a",
+  "candal": "candal_a",
+  "cantarell": "cantarell_a",
+  "cantata one": "cantata_one_a",
+  "cantora one": "cantora_one_a",
+  "caprasimo": "caprasimo_a",
+  "capriola": "capriola_a",
+  "caramel": "caramel_a",
+  "carattere": "carattere_a",
+  "cardo": "cardo_a",
+  "carlito": "carlito_a",
+  "carme": "carme_a",
+  "carrois gothic": "carrois_gothic_a",
+  "carrois gothic sc": "carrois_gothic_sc_a",
+  "carter one": "carter_one_a",
+  "cascadia code": "cascadia_code_a",
+  "cascadia mono": "cascadia_mono_a",
+  "castoro": "castoro_a",
+  "castoro titling": "castoro_titling_a",
+  "catamaran": "catamaran_a",
+  "caveat": "caveat_a",
+  "caveat brush": "caveat_brush_a",
+  "cedarville cursive": "cedarville_cursive_a",
+  "ceviche one": "ceviche_one_a",
+  "chakra petch": "chakra_petch_a"
+}

data/formatting_filters.csv

@@ -0,0 +1,1201 @@
+run,step,metric,value,stderr
+Baseline,1000,ai2d_exact_match,0.2548575129533679,0.007843322436924496
+Baseline,1000,average,0.27120689295763617,
+Baseline,1000,average_rank,3.8,
+Baseline,1000,chartqa_relaxed_overall,0.3308,0.009411906161401973
+Baseline,1000,docvqa_val_anls,0.3528553494243383,0.005852289239342309
+Baseline,1000,infovqa_val_anls,0.17320578642581314,0.006297063452679795
+Baseline,1000,mme_total_score,977.4280712284914,
+Baseline,1000,mmmu_val_mmmu_acc,0.25222,
+Baseline,1000,mmstar_average,0.23215874078908072,
+Baseline,1000,ocrbench_ocrbench_accuracy,0.286,
+Baseline,1000,seedbench_seed_all,0.2563646470261256,
+Baseline,1000,textvqa_val_exact_match,0.3024,0.00628900296642181
+Baseline,2000,ai2d_exact_match,0.26295336787564766,0.007923526907377255
+Baseline,2000,average,0.3202068275596269,
+Baseline,2000,average_rank,3.7,
+Baseline,2000,chartqa_relaxed_overall,0.4688,0.009982508912777261
+Baseline,2000,docvqa_val_anls,0.4452261510942785,0.00614755494712251
+Baseline,2000,infovqa_val_anls,0.1820547866557169,0.006217861455795791
+Baseline,2000,mme_total_score,1049.3036214485794,
+Baseline,2000,mmmu_val_mmmu_acc,0.24556,
+Baseline,2000,mmstar_average,0.21305462434540698,
+Baseline,2000,ocrbench_ocrbench_accuracy,0.395,
+Baseline,2000,seedbench_seed_all,0.258532518065592,
+Baseline,2000,textvqa_val_exact_match,0.41068000000000005,0.006697862330024289
+Baseline,3000,ai2d_exact_match,0.25226683937823835,0.007816909588794397
+Baseline,3000,average,0.3507423834414229,
+Baseline,3000,average_rank,2.6,
+Baseline,3000,chartqa_relaxed_overall,0.5028,0.010001843767601082
+Baseline,3000,docvqa_val_anls,0.502653993831009,0.006267072346683124
+Baseline,3000,infovqa_val_anls,0.21728617578189535,0.006796941784959762
+Baseline,3000,mme_total_score,1170.2383953581434,
+Baseline,3000,mmmu_val_mmmu_acc,0.27556,
+Baseline,3000,mmstar_average,0.25432376938577683,
+Baseline,3000,ocrbench_ocrbench_accuracy,0.436,
+Baseline,3000,seedbench_seed_all,0.2792106725958866,
+Baseline,3000,textvqa_val_exact_match,0.43658,0.006766885462882726
+Baseline,4000,ai2d_exact_match,0.2645725388601036,0.007939149662089447
+Baseline,4000,average,0.36961781722974835,
+Baseline,4000,average_rank,2.8,
+Baseline,4000,chartqa_relaxed_overall,0.5312,0.009982508912777261
+Baseline,4000,docvqa_val_anls,0.5374434618615119,0.0062905728113059655
+Baseline,4000,infovqa_val_anls,0.2287924838861707,0.006994568698639919
+Baseline,4000,mme_total_score,1155.203781512605,
+Baseline,4000,mmmu_val_mmmu_acc,0.25556,
+Baseline,4000,mmstar_average,0.2575590188757354,
+Baseline,4000,ocrbench_ocrbench_accuracy,0.453,
+Baseline,4000,seedbench_seed_all,0.33913285158421347,
+Baseline,4000,textvqa_val_exact_match,0.4593,0.006791695475025738
+Baseline,5000,ai2d_exact_match,0.3125,0.008342439145556371
+Baseline,5000,average,0.3974627910380972,
+Baseline,5000,average_rank,3.1,
+Baseline,5000,chartqa_relaxed_overall,0.5488,0.00995424828018316
+Baseline,5000,docvqa_val_anls,0.552360266782429,0.006300308519952055
+Baseline,5000,infovqa_val_anls,0.23425555286643698,0.007002254622066442
+Baseline,5000,mme_total_score,1181.4653861544618,
+Baseline,5000,mmmu_val_mmmu_acc,0.26667,
+Baseline,5000,mmstar_average,0.29596648146165705,
+Baseline,5000,ocrbench_ocrbench_accuracy,0.462,
+Baseline,5000,seedbench_seed_all,0.43107281823235133,
+Baseline,5000,textvqa_val_exact_match,0.47354000000000007,0.0068172185364497985
+Baseline,6000,ai2d_exact_match,0.358160621761658,0.008629463221867162
+Baseline,6000,average,0.4161227404571003,
+Baseline,6000,average_rank,2.3,
+Baseline,6000,chartqa_relaxed_overall,0.5628,0.00992279440175477
+Baseline,6000,docvqa_val_anls,0.5747451497228876,0.00625495440870239
+Baseline,6000,infovqa_val_anls,0.22152017368968838,0.006604546680525351
+Baseline,6000,mme_total_score,1284.1648659463785,
+Baseline,6000,mmmu_val_mmmu_acc,0.27111,
+Baseline,6000,mmstar_average,0.2978489412854164,
+Baseline,6000,ocrbench_ocrbench_accuracy,0.495,
+Baseline,6000,seedbench_seed_all,0.4795997776542524,
+Baseline,6000,textvqa_val_exact_match,0.48432,0.006800535050670284
+Baseline,7000,ai2d_exact_match,0.3707901554404145,0.00869347755587734
+Baseline,7000,average,0.4291083177345374,
+Baseline,7000,average_rank,2.6,
+Baseline,7000,chartqa_relaxed_overall,0.5656,0.009915542506251351
+Baseline,7000,docvqa_val_anls,0.5940907049431567,0.006224236305767187
+Baseline,7000,infovqa_val_anls,0.2515675215816963,0.007105097396092786
+Baseline,7000,mme_total_score,1185.875650260104,
+Baseline,7000,mmmu_val_mmmu_acc,0.26556,
+Baseline,7000,mmstar_average,0.31372400960777047,
+Baseline,7000,ocrbench_ocrbench_accuracy,0.504,
+Baseline,7000,seedbench_seed_all,0.4964424680377988,
+Baseline,7000,textvqa_val_exact_match,0.5002,0.006794794025220267
+Baseline,8000,ai2d_exact_match,0.37759067357512954,0.008725299846043883
+Baseline,8000,average,0.43846759477995995,
+Baseline,8000,average_rank,2.1,
+Baseline,8000,chartqa_relaxed_overall,0.5832,0.009862556058385773
+Baseline,8000,docvqa_val_anls,0.6017336419437208,0.006231612198089698
+Baseline,8000,infovqa_val_anls,0.2449256624147254,0.006992518502948913
+Baseline,8000,mme_total_score,1199.2409963985594,
+Baseline,8000,mmmu_val_mmmu_acc,0.28111,
+Baseline,8000,mmstar_average,0.33512257186205047,
+Baseline,8000,ocrbench_ocrbench_accuracy,0.51,
+Baseline,8000,seedbench_seed_all,0.5024458032240133,
+Baseline,8000,textvqa_val_exact_match,0.51008,0.006796301690135059
+Baseline,9000,ai2d_exact_match,0.4067357512953368,0.008841214921078996
+Baseline,9000,average,0.4422510732201056,
+Baseline,9000,average_rank,2.1,
+Baseline,9000,chartqa_relaxed_overall,0.5912,0.009834211136815875
+Baseline,9000,docvqa_val_anls,0.6170968481662739,0.00617235763542544
+Baseline,9000,infovqa_val_anls,0.23537031288570615,0.00670318154156447
+Baseline,9000,mme_total_score,1231.5195078031213,
+Baseline,9000,mmmu_val_mmmu_acc,0.25889,
+Baseline,9000,mmstar_average,0.3216444898242951,
+Baseline,9000,ocrbench_ocrbench_accuracy,0.515,
+Baseline,9000,seedbench_seed_all,0.5120622568093385,
+Baseline,9000,textvqa_val_exact_match,0.52226,0.006792711289708482
+Baseline,10000,ai2d_exact_match,0.39993523316062174,0.008817096257082848
+Baseline,10000,average,0.4523875703250908,
+Baseline,10000,average_rank,2.4,
+Baseline,10000,chartqa_relaxed_overall,0.5996,0.00980154906867574
+Baseline,10000,docvqa_val_anls,0.6262613496433054,0.006147756371688175
+Baseline,10000,infovqa_val_anls,0.263290074230132,0.007186788766942786
+Baseline,10000,mme_total_score,1240.8218287314926,
+Baseline,10000,mmmu_val_mmmu_acc,0.28778,
+Baseline,10000,mmstar_average,0.32972717906018517,
+Baseline,10000,ocrbench_ocrbench_accuracy,0.517,
+Baseline,10000,seedbench_seed_all,0.5217342968315731,
+Baseline,10000,textvqa_val_exact_match,0.5261600000000001,0.006785774843600811
+Baseline,11000,ai2d_exact_match,0.422279792746114,0.008889771831066474
+Baseline,11000,average,0.4561398159525099,
+Baseline,11000,average_rank,2.1,
+Baseline,11000,chartqa_relaxed_overall,0.6104,0.009755142291143075
+Baseline,11000,docvqa_val_anls,0.6373130149166712,0.006128022584995044
+Baseline,11000,infovqa_val_anls,0.24419378339723755,0.006897644885887063
+Baseline,11000,mme_total_score,1322.9488795518205,
+Baseline,11000,mmmu_val_mmmu_acc,0.27778,
+Baseline,11000,mmstar_average,0.3298563439522548,
+Baseline,11000,ocrbench_ocrbench_accuracy,0.521,
+Baseline,11000,seedbench_seed_all,0.5237354085603113,
+Baseline,11000,textvqa_val_exact_match,0.5387,0.006770851562852138
+Baseline,12000,ai2d_exact_match,0.42001295336787564,0.008883255931688034
+Baseline,12000,average,0.4582751140055433,
+Baseline,12000,average_rank,2.4,
+Baseline,12000,chartqa_relaxed_overall,0.618,0.009719474639861454
+Baseline,12000,docvqa_val_anls,0.6393961983751871,0.0061228747388476674
+Baseline,12000,infovqa_val_anls,0.24798874058574302,0.006855374548993139
+Baseline,12000,mme_total_score,1225.6453581432572,
+Baseline,12000,mmmu_val_mmmu_acc,0.27889,
+Baseline,12000,mmstar_average,0.34010867846816534,
+Baseline,12000,ocrbench_ocrbench_accuracy,0.512,
+Baseline,12000,seedbench_seed_all,0.5350194552529183,
+Baseline,12000,textvqa_val_exact_match,0.5330600000000001,0.006777713092109446
+Baseline,13000,ai2d_exact_match,0.4375,0.008928571428571428
+Baseline,13000,average,0.4692868662590049,
+Baseline,13000,average_rank,1.6,
+Baseline,13000,chartqa_relaxed_overall,0.6148,0.00973479791861169
+Baseline,13000,docvqa_val_anls,0.6511374872549951,0.006086953065248391
+Baseline,13000,infovqa_val_anls,0.24465055100441893,0.006808432538374664
+Baseline,13000,mme_total_score,1281.7122849139657,
+Baseline,13000,mmmu_val_mmmu_acc,0.28222,
+Baseline,13000,mmstar_average,0.3453069542917521,
+Baseline,13000,ocrbench_ocrbench_accuracy,0.549,
+Baseline,13000,seedbench_seed_all,0.5442468037798777,
+Baseline,13000,textvqa_val_exact_match,0.55472,0.0067416788982325
+Baseline,14000,ai2d_exact_match,0.4572538860103627,0.00896620675297095
+Baseline,14000,average,0.47352486841689195,
+Baseline,14000,average_rank,1.7,
+Baseline,14000,chartqa_relaxed_overall,0.6172,0.009723347231923635
+Baseline,14000,docvqa_val_anls,0.6502269393708169,0.006057950730638126
+Baseline,14000,infovqa_val_anls,0.25805460837190913,0.007037735231659539
+Baseline,14000,mme_total_score,1309.1444577831132,
+Baseline,14000,mmmu_val_mmmu_acc,0.28111,
+Baseline,14000,mmstar_average,0.34575818188776586,
+Baseline,14000,ocrbench_ocrbench_accuracy,0.551,
+Baseline,14000,seedbench_seed_all,0.5483602001111729,
+Baseline,14000,textvqa_val_exact_match,0.55276,0.006751206724612103
+Baseline,15000,ai2d_exact_match,0.45045336787564766,0.008954861634252399
+Baseline,15000,average,0.47878665012878824,
+Baseline,15000,average_rank,1.6,
+Baseline,15000,chartqa_relaxed_overall,0.612,0.009747841205275417
+Baseline,15000,docvqa_val_anls,0.6621413031955148,0.006056838050222495
+Baseline,15000,infovqa_val_anls,0.2706898598157733,0.007200315730154543
+Baseline,15000,mme_total_score,1384.2171868747498,
+Baseline,15000,mmmu_val_mmmu_acc,0.30222,
+Baseline,15000,mmstar_average,0.35408135695920684,
+Baseline,15000,ocrbench_ocrbench_accuracy,0.558,
+Baseline,15000,seedbench_seed_all,0.5411339633129516,
+Baseline,15000,textvqa_val_exact_match,0.5583600000000001,0.0067279027203879065
+Baseline,16000,ai2d_exact_match,0.45077720207253885,0.008955440137395838
+Baseline,16000,average,0.47665128022935843,
+Baseline,16000,average_rank,1.6,
+Baseline,16000,chartqa_relaxed_overall,0.632,0.00964715642305132
+Baseline,16000,docvqa_val_anls,0.6709415729142987,0.005999818105621502
+Baseline,16000,infovqa_val_anls,0.26050032542402035,0.006997451875879188
+Baseline,16000,mme_total_score,1317.8491396558625,
+Baseline,16000,mmmu_val_mmmu_acc,0.27556,
+Baseline,16000,mmstar_average,0.33214333327093315,
+Baseline,16000,ocrbench_ocrbench_accuracy,0.56,
+Baseline,16000,seedbench_seed_all,0.5463590883824346,
+Baseline,16000,textvqa_val_exact_match,0.56158,0.006723854754867398
+Baseline,17000,ai2d_exact_match,0.45919689119170987,0.008969138793675545
+Baseline,17000,average,0.4777141780162423,
+Baseline,17000,average_rank,1.9,
+Baseline,17000,chartqa_relaxed_overall,0.632,0.00964715642305132
+Baseline,17000,docvqa_val_anls,0.6796338519136422,0.005948761388267941
+Baseline,17000,infovqa_val_anls,0.28070956072505215,0.007298333094144192
+Baseline,17000,mme_total_score,1381.9161664665867,
+Baseline,17000,mmmu_val_mmmu_acc,0.27667,
+Baseline,17000,mmstar_average,0.3370289492329521,
+Baseline,17000,ocrbench_ocrbench_accuracy,0.519,
+Baseline,17000,seedbench_seed_all,0.5510283490828238,
+Baseline,17000,textvqa_val_exact_match,0.56416,0.006724830373229479
+Baseline,18000,ai2d_exact_match,0.46567357512953367,0.008977921602780726
+Baseline,18000,average,0.4819834595278701,
+Baseline,18000,average_rank,1.7,
+Baseline,18000,chartqa_relaxed_overall,0.6376,0.009615793331418735
+Baseline,18000,docvqa_val_anls,0.6775884603912571,0.005972234236435759
+Baseline,18000,infovqa_val_anls,0.27154318420389256,0.007164903131667027
+Baseline,18000,mme_total_score,1336.922769107643,
+Baseline,18000,mmmu_val_mmmu_acc,0.28667,
+Baseline,18000,mmstar_average,0.34482796716566916,
+Baseline,18000,ocrbench_ocrbench_accuracy,0.533,
+Baseline,18000,seedbench_seed_all,0.5543079488604781,
+Baseline,18000,textvqa_val_exact_match,0.5666399999999999,0.006713392287599574
+Baseline,19000,ai2d_exact_match,0.4682642487046632,0.008981008686994101
+Baseline,19000,average,0.4899006713916878,
+Baseline,19000,average_rank,1.5,
+Baseline,19000,chartqa_relaxed_overall,0.6444,0.009575809858898698
+Baseline,19000,docvqa_val_anls,0.678226526479947,0.005970619221588814
+Baseline,19000,infovqa_val_anls,0.26993847247278,0.0071348470764911525
+Baseline,19000,mme_total_score,1406.6628651460583,
+Baseline,19000,mmmu_val_mmmu_acc,0.28333,
+Baseline,19000,mmstar_average,0.356220913822775,
+Baseline,19000,ocrbench_ocrbench_accuracy,0.577,
+Baseline,19000,seedbench_seed_all,0.554585881045025,
+Baseline,19000,textvqa_val_exact_match,0.57714,0.0066918487914812905
+Baseline,20000,ai2d_exact_match,0.47571243523316065,0.00898853090258662
+Baseline,20000,average,0.4873169067639118,
+Baseline,20000,average_rank,1.7,
+Baseline,20000,chartqa_relaxed_overall,0.6336,0.009638338810708618
+Baseline,20000,docvqa_val_anls,0.6895214454380043,0.005896462073053767
+Baseline,20000,infovqa_val_anls,0.2655657550458317,0.007033265532032538
+Baseline,20000,mme_total_score,1324.6738695478193,
+Baseline,20000,mmmu_val_mmmu_acc,0.30111,
+Baseline,20000,mmstar_average,0.33806766134497995,
+Baseline,20000,ocrbench_ocrbench_accuracy,0.555,
+Baseline,20000,seedbench_seed_all,0.5587548638132296,
+Baseline,20000,textvqa_val_exact_match,0.56852,0.006720151338087659
+≥2,1000,ai2d_exact_match,0.2619818652849741,0.007914086941902845
+≥2,1000,average,0.2852885776543714,
+≥2,1000,average_rank,2.8,
+≥2,1000,chartqa_relaxed_overall,0.36,0.009601920576192066
+≥2,1000,docvqa_val_anls,0.3691495236959511,0.0059102400877721764
+≥2,1000,infovqa_val_anls,0.18005913830944342,0.006300821228003093
+≥2,1000,mme_total_score,1034.4992997198879,
+≥2,1000,mmmu_val_mmmu_acc,0.25222,
+≥2,1000,mmstar_average,0.20333316409480473,
+≥2,1000,ocrbench_ocrbench_accuracy,0.331,
+≥2,1000,seedbench_seed_all,0.264313507504169,
+≥2,1000,textvqa_val_exact_match,0.34554,0.006483180392801138
+≥2,2000,ai2d_exact_match,0.25971502590673573,0.007891865786132407
+≥2,2000,average,0.3309474525195546,
+≥2,2000,average_rank,2.3,
+≥2,2000,chartqa_relaxed_overall,0.4664,0.009979391329160321
+≥2,2000,docvqa_val_anls,0.44591152951098784,0.006197056264354256
+≥2,2000,infovqa_val_anls,0.20775747304541303,0.006645281501282388
+≥2,2000,mme_total_score,1083.1982793117247,
+≥2,2000,mmmu_val_mmmu_acc,0.26222,
+≥2,2000,mmstar_average,0.23515873070535015,
+≥2,2000,ocrbench_ocrbench_accuracy,0.413,
+≥2,2000,seedbench_seed_all,0.2757643135075042,
+≥2,2000,textvqa_val_exact_match,0.4126,0.006707581257746032
+≥2,3000,ai2d_exact_match,0.27299222797927464,0.00801819019286542
+≥2,3000,average,0.35386512749374127,
+≥2,3000,average_rank,1.8,
+≥2,3000,chartqa_relaxed_overall,0.5124,0.009998924311892653
+≥2,3000,docvqa_val_anls,0.48910828732243933,0.006274136020264289
+≥2,3000,infovqa_val_anls,0.2070472808493129,0.0065577848697521875
+≥2,3000,mme_total_score,1128.4556822729091,
+≥2,3000,mmmu_val_mmmu_acc,0.25222,
+≥2,3000,mmstar_average,0.25601322622316125,
+≥2,3000,ocrbench_ocrbench_accuracy,0.447,
+≥2,3000,seedbench_seed_all,0.30522512506948307,
+≥2,3000,textvqa_val_exact_match,0.44278,0.006763750733490772
+≥2,4000,ai2d_exact_match,0.3173575129533679,0.008377274276497445
+≥2,4000,average,0.3859767914191833,
+≥2,4000,average_rank,2.4,
+≥2,4000,chartqa_relaxed_overall,0.5388,0.0099718403035556
+≥2,4000,docvqa_val_anls,0.5360750144064242,0.006293888693576319
+≥2,4000,infovqa_val_anls,0.20188347210673038,0.0064101781168258935
+≥2,4000,mme_total_score,1110.1481592637056,
+≥2,4000,mmmu_val_mmmu_acc,0.25222,
+≥2,4000,mmstar_average,0.28905163247788923,
+≥2,4000,ocrbench_ocrbench_accuracy,0.473,
+≥2,4000,seedbench_seed_all,0.4102834908282379,
+≥2,4000,textvqa_val_exact_match,0.4551200000000001,0.0067829756649846785
+≥2,5000,ai2d_exact_match,0.32577720207253885,0.008435168191407938
+≥2,5000,average,0.4029564015858402,
+≥2,5000,average_rank,2.5,
+≥2,5000,chartqa_relaxed_overall,0.5628,0.00992279440175477
+≥2,5000,docvqa_val_anls,0.548770508666009,0.006315482288099859
+≥2,5000,infovqa_val_anls,0.20783386531525747,0.006421967027729742
+≥2,5000,mme_total_score,1206.0127050820329,
+≥2,5000,mmmu_val_mmmu_acc,0.25667,
+≥2,5000,mmstar_average,0.3210115801865161,
+≥2,5000,ocrbench_ocrbench_accuracy,0.484,
+≥2,5000,seedbench_seed_all,0.4440244580322401,
+≥2,5000,textvqa_val_exact_match,0.47572,0.006783457774606987
+≥2,6000,ai2d_exact_match,0.3542746113989637,0.00860846328571982
+≥2,6000,average,0.4118759304334577,
+≥2,6000,average_rank,3.3,
+≥2,6000,chartqa_relaxed_overall,0.5644,0.00991868984106597
+≥2,6000,docvqa_val_anls,0.5618652265799138,0.006261889040657647
+≥2,6000,infovqa_val_anls,0.2101901707833487,0.006387610514125727
+≥2,6000,mme_total_score,1135.471288515406,
+≥2,6000,mmmu_val_mmmu_acc,0.26333,
+≥2,6000,mmstar_average,0.3255666447386709,
+≥2,6000,ocrbench_ocrbench_accuracy,0.482,
+≥2,6000,seedbench_seed_all,0.4740967204002223,
+≥2,6000,textvqa_val_exact_match,0.47116,0.00678908456375694
+≥2,7000,ai2d_exact_match,0.37338082901554404,0.008705816961084268
+≥2,7000,average,0.4291995483001856,
+≥2,7000,average_rank,2.4,
+≥2,7000,chartqa_relaxed_overall,0.5716,0.009898917689756362
+≥2,7000,docvqa_val_anls,0.5846126379804475,0.006218823793449337
+≥2,7000,infovqa_val_anls,0.2243908724169204,0.006651785538916188
+≥2,7000,mme_total_score,1249.9180672268908,
+≥2,7000,mmmu_val_mmmu_acc,0.27556,
+≥2,7000,mmstar_average,0.32644844909642934,
+≥2,7000,ocrbench_ocrbench_accuracy,0.506,
+≥2,7000,seedbench_seed_all,0.4936631461923291,
+≥2,7000,textvqa_val_exact_match,0.5071399999999999,0.00678246300696791
+≥2,8000,ai2d_exact_match,0.3963730569948187,0.008803757198545703
+≥2,8000,average,0.43448504857720643,
+≥2,8000,average_rank,3.0,
+≥2,8000,chartqa_relaxed_overall,0.5784,0.009878279615563902
+≥2,8000,docvqa_val_anls,0.5935884981677085,0.006228109848938283
+≥2,8000,infovqa_val_anls,0.22034669568379356,0.006538842004996925
+≥2,8000,mme_total_score,1251.6327531012405,
+≥2,8000,mmmu_val_mmmu_acc,0.27444,
+≥2,8000,mmstar_average,0.3368503047476477,
+≥2,8000,ocrbench_ocrbench_accuracy,0.516,
+≥2,8000,seedbench_seed_all,0.4963868816008894,
+≥2,8000,textvqa_val_exact_match,0.49798000000000003,0.006777844181917349
+≥2,9000,ai2d_exact_match,0.4015544041450777,0.008822998789014784
+≥2,9000,average,0.4409076865862069,
+≥2,9000,average_rank,2.8,
+≥2,9000,chartqa_relaxed_overall,0.5952,0.0098190299592035
+≥2,9000,docvqa_val_anls,0.6142957639909281,0.006149142953850004
+≥2,9000,infovqa_val_anls,0.225441641847203,0.006565814507342015
+≥2,9000,mme_total_score,1170.923569427771,
+≥2,9000,mmmu_val_mmmu_acc,0.28,
+≥2,9000,mmstar_average,0.32763888124373686,
+≥2,9000,ocrbench_ocrbench_accuracy,0.514,
+≥2,9000,seedbench_seed_all,0.5012784880489161,
+≥2,9000,textvqa_val_exact_match,0.50876,0.006788539558245703
+≥2,10000,ai2d_exact_match,0.4018782383419689,0.008824167272304229
+≥2,10000,average,0.44844067183729286,
+≥2,10000,average_rank,2.7,
+≥2,10000,chartqa_relaxed_overall,0.5956,0.009817474681589429
+≥2,10000,docvqa_val_anls,0.6161881255627961,0.006150295182189919
+≥2,10000,infovqa_val_anls,0.2273186020139702,0.006609762944776786
+≥2,10000,mme_total_score,1244.6918767507002,
+≥2,10000,mmmu_val_mmmu_acc,0.28667,
+≥2,10000,mmstar_average,0.3405769394273513,
+≥2,10000,ocrbench_ocrbench_accuracy,0.529,
+≥2,10000,seedbench_seed_all,0.5174541411895498,
+≥2,10000,textvqa_val_exact_match,0.52128,0.0067723707312184415
+≥2,11000,ai2d_exact_match,0.41386010362694303,0.008864599272573477
+≥2,11000,average,0.4508015001273205,
+≥2,11000,average_rank,3.1,
+≥2,11000,chartqa_relaxed_overall,0.5916,0.0098327233755248
+≥2,11000,docvqa_val_anls,0.6132516406541649,0.006147223601932411
+≥2,11000,infovqa_val_anls,0.23136501765139353,0.006670154065298524
+≥2,11000,mme_total_score,1193.1198479391755,
+≥2,11000,mmmu_val_mmmu_acc,0.28222,
+≥2,11000,mmstar_average,0.34055130285985363,
+≥2,11000,ocrbench_ocrbench_accuracy,0.544,
+≥2,11000,seedbench_seed_all,0.5137854363535297,
+≥2,11000,textvqa_val_exact_match,0.52658,0.006779520123033763
+≥2,12000,ai2d_exact_match,0.42033678756476683,0.008884198538329101
+≥2,12000,average,0.4593162089992856,
+≥2,12000,average_rank,2.4,
+≥2,12000,chartqa_relaxed_overall,0.612,0.009747841205275417
+≥2,12000,docvqa_val_anls,0.6322256818549263,0.006037251396803284
+≥2,12000,infovqa_val_anls,0.23499854511160906,0.006635085630122106
+≥2,12000,mme_total_score,1282.3226290516207,
+≥2,12000,mmmu_val_mmmu_acc,0.29444,
+≥2,12000,mmstar_average,0.3455632878074604,
+≥2,12000,ocrbench_ocrbench_accuracy,0.542,
+≥2,12000,seedbench_seed_all,0.5148415786548082,
+≥2,12000,textvqa_val_exact_match,0.5374399999999999,0.0067549667056943374
+≥2,13000,ai2d_exact_match,0.4329663212435233,0.008917911748577596
+≥2,13000,average,0.4594856750450977,
+≥2,13000,average_rank,3.1,
+≥2,13000,chartqa_relaxed_overall,0.6116,0.009749676839741497
+≥2,13000,docvqa_val_anls,0.6480115225202001,0.006082136258345928
+≥2,13000,infovqa_val_anls,0.2390399772273204,0.006801403608154099
+≥2,13000,mme_total_score,1255.4888955582232,
+≥2,13000,mmmu_val_mmmu_acc,0.26667,
+≥2,13000,mmstar_average,0.3276926929918222,
+≥2,13000,ocrbench_ocrbench_accuracy,0.551,
+≥2,13000,seedbench_seed_all,0.5190105614230128,
+≥2,13000,textvqa_val_exact_match,0.5393800000000001,0.006748937157104821
+≥2,14000,ai2d_exact_match,0.43523316062176165,0.00892333645202351
+≥2,14000,average,0.46380397688227554,
+≥2,14000,average_rank,3.4,
+≥2,14000,chartqa_relaxed_overall,0.6136,0.009740429476494075
+≥2,14000,docvqa_val_anls,0.6474419557198757,0.006056802443739013
+≥2,14000,infovqa_val_anls,0.24341248035748822,0.006789396426159645
+≥2,14000,mme_total_score,1209.5489195678272,
+≥2,14000,mmmu_val_mmmu_acc,0.27556,
+≥2,14000,mmstar_average,0.35309886783724065,
+≥2,14000,ocrbench_ocrbench_accuracy,0.545,
+≥2,14000,seedbench_seed_all,0.5207893274041134,
+≥2,14000,textvqa_val_exact_match,0.5400999999999999,0.006762835587905254
+≥2,15000,ai2d_exact_match,0.43588082901554404,0.008924851504668983
+≥2,15000,average,0.46327247775474995,
+≥2,15000,average_rank,3.4,
+≥2,15000,chartqa_relaxed_overall,0.614,0.009738559226822298
+≥2,15000,docvqa_val_anls,0.638973421646662,0.005999307255506728
+≥2,15000,infovqa_val_anls,0.23590457960067904,0.006699424952743598
+≥2,15000,mme_total_score,1230.12775110044,
+≥2,15000,mmmu_val_mmmu_acc,0.28667,
+≥2,15000,mmstar_average,0.3545309625815601,
+≥2,15000,ocrbench_ocrbench_accuracy,0.536,
+≥2,15000,seedbench_seed_all,0.5225125069483046,
+≥2,15000,textvqa_val_exact_match,0.54498,0.006749227387936104
+≥2,16000,ai2d_exact_match,0.4413860103626943,0.008937105222785166
+≥2,16000,average,0.4691799993543692,
+≥2,16000,average_rank,3.1,
+≥2,16000,chartqa_relaxed_overall,0.6168,0.009725273074549106
+≥2,16000,docvqa_val_anls,0.6539654303329543,0.00605387835402321
+≥2,16000,infovqa_val_anls,0.251011584102177,0.006888371171829252
+≥2,16000,mme_total_score,1235.6986794717886,
+≥2,16000,mmmu_val_mmmu_acc,0.28556,
+≥2,16000,mmstar_average,0.35467603553935745,
+≥2,16000,ocrbench_ocrbench_accuracy,0.545,
+≥2,16000,seedbench_seed_all,0.5256809338521401,
+≥2,16000,textvqa_val_exact_match,0.5485399999999999,0.0067546057338473825
+≥2,17000,ai2d_exact_match,0.43976683937823835,0.008933617011753861
+≥2,17000,average,0.47032074037035837,
+≥2,17000,average_rank,3.1,
+≥2,17000,chartqa_relaxed_overall,0.6184,0.009717527882093043
+≥2,17000,docvqa_val_anls,0.655318828070759,0.005978737407680595
+≥2,17000,infovqa_val_anls,0.24899610305034758,0.006825123869520012
+≥2,17000,mme_total_score,1246.5548219287716,
+≥2,17000,mmmu_val_mmmu_acc,0.29556,
+≥2,17000,mmstar_average,0.3399003792152031,
+≥2,17000,ocrbench_ocrbench_accuracy,0.553,
+≥2,17000,seedbench_seed_all,0.5241245136186771,
+≥2,17000,textvqa_val_exact_match,0.55782,0.0067278542139723035
+≥2,18000,ai2d_exact_match,0.4413860103626943,0.008937105222785166
+≥2,18000,average,0.4720458439616472,
+≥2,18000,average_rank,3.0,
+≥2,18000,chartqa_relaxed_overall,0.6256,0.009681288495793083
+≥2,18000,docvqa_val_anls,0.6595541124701471,0.00598982698063352
+≥2,18000,infovqa_val_anls,0.24628476636774824,0.006771852338911992
+≥2,18000,mme_total_score,1246.580632252901,
+≥2,18000,mmmu_val_mmmu_acc,0.29556,
+≥2,18000,mmstar_average,0.3370877286888103,
+≥2,18000,ocrbench_ocrbench_accuracy,0.557,
+≥2,18000,seedbench_seed_all,0.5279599777654252,
+≥2,18000,textvqa_val_exact_match,0.5579800000000001,0.006730967620262408
+≥2,19000,ai2d_exact_match,0.44009067357512954,0.008934322367529354
+≥2,19000,average,0.47484048232505544,
+≥2,19000,average_rank,3.1,
+≥2,19000,chartqa_relaxed_overall,0.6236,0.009691583292459796
+≥2,19000,docvqa_val_anls,0.656845810830509,0.005998201285366962
+≥2,19000,infovqa_val_anls,0.2483428639081206,0.006859079557818024
+≥2,19000,mme_total_score,1255.002801120448,
+≥2,19000,mmmu_val_mmmu_acc,0.30667,
+≥2,19000,mmstar_average,0.3484268658746636,
+≥2,19000,ocrbench_ocrbench_accuracy,0.557,
+≥2,19000,seedbench_seed_all,0.5306281267370762,
+≥2,19000,textvqa_val_exact_match,0.56196,0.006711810587335734
+≥2,20000,ai2d_exact_match,0.44397668393782386,0.008942485993062323
+≥2,20000,average,0.4719647885204447,
+≥2,20000,average_rank,3.3,
+≥2,20000,chartqa_relaxed_overall,0.6252,0.009683361554563506
+≥2,20000,docvqa_val_anls,0.6531065052301426,0.005958657790556006
+≥2,20000,infovqa_val_anls,0.2515640311557441,0.00684713602725156
+≥2,20000,mme_total_score,1269.56512605042,
+≥2,20000,mmmu_val_mmmu_acc,0.29222,
+≥2,20000,mmstar_average,0.3405247257210479,
+≥2,20000,ocrbench_ocrbench_accuracy,0.557,
+≥2,20000,seedbench_seed_all,0.528071150639244,
+≥2,20000,textvqa_val_exact_match,0.5560200000000001,0.006742124529303335
+≥3,1000,ai2d_exact_match,0.25712435233160624,0.007866134203324925
+≥3,1000,average,0.2908366935977347,
+≥3,1000,average_rank,2.6,
+≥3,1000,chartqa_relaxed_overall,0.3724,0.009670817229291067
+≥3,1000,docvqa_val_anls,0.36190361730095816,0.005874681377878617
+≥3,1000,infovqa_val_anls,0.1897409167650202,0.006570751118077319
+≥3,1000,mme_total_score,938.3572428971589,
+≥3,1000,mmmu_val_mmmu_acc,0.26333,
+≥3,1000,mmstar_average,0.2361438073438948,
+≥3,1000,ocrbench_ocrbench_accuracy,0.313,
+≥3,1000,seedbench_seed_all,0.25758754863813227,
+≥3,1000,textvqa_val_exact_match,0.36629999999999996,0.006582851113746775
+≥3,2000,ai2d_exact_match,0.25647668393782386,0.007859644922870102
+≥3,2000,average,0.327478691314146,
+≥3,2000,average_rank,2.9,
+≥3,2000,chartqa_relaxed_overall,0.4708,0.009984929820955767
+≥3,2000,docvqa_val_anls,0.455859181323049,0.0061669106819143196
+≥3,2000,infovqa_val_anls,0.20804914764579785,0.0066905821266465505
+≥3,2000,mme_total_score,990.4248699479792,
+≥3,2000,mmmu_val_mmmu_acc,0.27111,
+≥3,2000,mmstar_average,0.21380673865938699,
+≥3,2000,ocrbench_ocrbench_accuracy,0.405,
+≥3,2000,seedbench_seed_all,0.26364647026125626,
+≥3,2000,textvqa_val_exact_match,0.40256,0.0066960030295180025
+≥3,3000,ai2d_exact_match,0.2697538860103627,0.007988222765138163
+≥3,3000,average,0.34640625458292296,
+≥3,3000,average_rank,3.1,
+≥3,3000,chartqa_relaxed_overall,0.514,0.009998079047189691
+≥3,3000,docvqa_val_anls,0.4749731938810012,0.00604931863100692
+≥3,3000,infovqa_val_anls,0.19785201580687228,0.00636819106561235
+≥3,3000,mme_total_score,1022.0748299319728,
+≥3,3000,mmmu_val_mmmu_acc,0.26556,
+≥3,3000,mmstar_average,0.2234035546364529,
+≥3,3000,ocrbench_ocrbench_accuracy,0.435,
+≥3,3000,seedbench_seed_all,0.29655364091161757,
+≥3,3000,textvqa_val_exact_match,0.44056000000000006,0.006770653264576898
+≥3,4000,ai2d_exact_match,0.31476683937823835,0.008358827401711809
+≥3,4000,average,0.3840989485719881,
+≥3,4000,average_rank,2.5,
+≥3,4000,chartqa_relaxed_overall,0.5244,0.009990083919101193
+≥3,4000,docvqa_val_anls,0.5398623644141017,0.006209437344747972
+≥3,4000,infovqa_val_anls,0.21841657659961455,0.006654701266433889
+≥3,4000,mme_total_score,1008.1938775510204,
+≥3,4000,mmmu_val_mmmu_acc,0.27778,
+≥3,4000,mmstar_average,0.25611290572758977,
+≥3,4000,ocrbench_ocrbench_accuracy,0.462,
+≥3,4000,seedbench_seed_all,0.39733185102834906,
+≥3,4000,textvqa_val_exact_match,0.46621999999999997,0.006799457981763631
+≥3,5000,ai2d_exact_match,0.3442357512953368,0.008551327504046387
+≥3,5000,average,0.4034839586685592,
+≥3,5000,average_rank,2.4,
+≥3,5000,chartqa_relaxed_overall,0.5544,0.009942625323290008
+≥3,5000,docvqa_val_anls,0.5567727758893183,0.006173642024037381
+≥3,5000,infovqa_val_anls,0.21638639427926507,0.006559084868006158
+≥3,5000,mme_total_score,1074.1284513805522,
+≥3,5000,mmmu_val_mmmu_acc,0.26778,
+≥3,5000,mmstar_average,0.3009278216170371,
+≥3,5000,ocrbench_ocrbench_accuracy,0.482,
+≥3,5000,seedbench_seed_all,0.4471928849360756,
+≥3,5000,textvqa_val_exact_match,0.46165999999999996,0.006793381991893107
+≥3,6000,ai2d_exact_match,0.36819948186528495,0.008680870162409787
+≥3,6000,average,0.41987173897944946,
+≥3,6000,average_rank,2.2,
+≥3,6000,chartqa_relaxed_overall,0.5636,0.009920755241100424
+≥3,6000,docvqa_val_anls,0.5766507662420887,0.006104661016322198
+≥3,6000,infovqa_val_anls,0.2209691160904877,0.0066290878786102805
+≥3,6000,mme_total_score,1088.6353541416568,
+≥3,6000,mmmu_val_mmmu_acc,0.29778,
+≥3,6000,mmstar_average,0.2960442855054549,
+≥3,6000,ocrbench_ocrbench_accuracy,0.496,
+≥3,6000,seedbench_seed_all,0.48360200111172874,
+≥3,6000,textvqa_val_exact_match,0.476,0.006791614329821814
+≥3,7000,ai2d_exact_match,0.3905440414507772,0.008780876258359173
+≥3,7000,average,0.4305333557001585,
+≥3,7000,average_rank,2.4,
+≥3,7000,chartqa_relaxed_overall,0.5744,0.009890651444389179
+≥3,7000,docvqa_val_anls,0.5943945047786826,0.006168637154272831
+≥3,7000,infovqa_val_anls,0.23015651757384684,0.006652654324068369
+≥3,7000,mme_total_score,1024.3486394557824,
+≥3,7000,mmmu_val_mmmu_acc,0.29,
+≥3,7000,mmstar_average,0.3086297067032336,
+≥3,7000,ocrbench_ocrbench_accuracy,0.496,
+≥3,7000,seedbench_seed_all,0.49577543079488606,
+≥3,7000,textvqa_val_exact_match,0.4949,0.006791673090238732
+≥3,8000,ai2d_exact_match,0.39863989637305697,0.008812301996070583
+≥3,8000,average,0.43894563539556736,
+≥3,8000,average_rank,2.4,
+≥3,8000,chartqa_relaxed_overall,0.5812,0.009869224115088964
+≥3,8000,docvqa_val_anls,0.597896936397571,0.006178924858305047
+≥3,8000,infovqa_val_anls,0.23624667779429379,0.006701812126185011
+≥3,8000,mme_total_score,1087.3003201280512,
+≥3,8000,mmmu_val_mmmu_acc,0.30444,
+≥3,8000,mmstar_average,0.3179238728089695,
+≥3,8000,ocrbench_ocrbench_accuracy,0.52,
+≥3,8000,seedbench_seed_all,0.5060033351862145,
+≥3,8000,textvqa_val_exact_match,0.48816000000000004,0.006805617250862191
+≥3,9000,ai2d_exact_match,0.4073834196891192,0.008843420154535594
+≥3,9000,average,0.4380819691649286,
+≥3,9000,average_rank,2.8,
+≥3,9000,chartqa_relaxed_overall,0.5892,0.009841548985529353
+≥3,9000,docvqa_val_anls,0.5926801961513722,0.00607014347283834
+≥3,9000,infovqa_val_anls,0.22884227739619317,0.006587321958723987
+≥3,9000,mme_total_score,960.1394557823129,
+≥3,9000,mmmu_val_mmmu_acc,0.29222,
+≥3,9000,mmstar_average,0.3023124740503409,
+≥3,9000,ocrbench_ocrbench_accuracy,0.516,
+≥3,9000,seedbench_seed_all,0.5108393551973318,
+≥3,9000,textvqa_val_exact_match,0.50326,0.006787480273097782
+≥3,10000,ai2d_exact_match,0.42487046632124353,0.008896983637113786
+≥3,10000,average,0.45376975700130806,
+≥3,10000,average_rank,2.5,
+≥3,10000,chartqa_relaxed_overall,0.592,0.009831228876620145
+≥3,10000,docvqa_val_anls,0.6288940533515488,0.006078026262812974
+≥3,10000,infovqa_val_anls,0.2639557991160976,0.007015193539901653
+≥3,10000,mme_total_score,1135.5116046418568,
+≥3,10000,mmmu_val_mmmu_acc,0.29556,
+≥3,10000,mmstar_average,0.3171165325775241,
+≥3,10000,ocrbench_ocrbench_accuracy,0.53,
+≥3,10000,seedbench_seed_all,0.5157309616453586,
+≥3,10000,textvqa_val_exact_match,0.5158,0.0067831610812991135
+≥3,11000,ai2d_exact_match,0.4271373056994819,0.008903088856242218
+≥3,11000,average,0.4507656942256156,
+≥3,11000,average_rank,3.1,
+≥3,11000,chartqa_relaxed_overall,0.6008,0.00979663889573671
+≥3,11000,docvqa_val_anls,0.6266233612884972,0.006097228164879785
+≥3,11000,infovqa_val_anls,0.23605295775343718,0.006674753327687541
+≥3,11000,mme_total_score,1115.0593237294918,
+≥3,11000,mmmu_val_mmmu_acc,0.27889,
+≥3,11000,mmstar_average,0.3244509807099135,
+≥3,11000,ocrbench_ocrbench_accuracy,0.524,
+≥3,11000,seedbench_seed_all,0.5219566425792107,
+≥3,11000,textvqa_val_exact_match,0.5169799999999999,0.006776837095888084
+≥3,12000,ai2d_exact_match,0.42843264248704666,0.008906491762178372
+≥3,12000,average,0.4596080978908205,
+≥3,12000,average_rank,2.6,
+≥3,12000,chartqa_relaxed_overall,0.6048,0.009779828322460816
+≥3,12000,docvqa_val_anls,0.6391083009950597,0.006038971765674556
+≥3,12000,infovqa_val_anls,0.24141834493583503,0.006794485284013245
+≥3,12000,mme_total_score,1183.3176270508202,
+≥3,12000,mmmu_val_mmmu_acc,0.28444,
+≥3,12000,mmstar_average,0.3293224419601992,
+≥3,12000,ocrbench_ocrbench_accuracy,0.555,
+≥3,12000,seedbench_seed_all,0.528071150639244,
+≥3,12000,textvqa_val_exact_match,0.5258799999999999,0.006773951756875811
+≥3,13000,ai2d_exact_match,0.43458549222797926,0.008921805911548515
+≥3,13000,average,0.4623863039639755,
+≥3,13000,average_rank,2.7,
+≥3,13000,chartqa_relaxed_overall,0.6108,0.00975332737879659
+≥3,13000,docvqa_val_anls,0.6376374898768016,0.006015671277879292
+≥3,13000,infovqa_val_anls,0.24710671614089955,0.006756961641692092
+≥3,13000,mme_total_score,1261.84493797519,
+≥3,13000,mmmu_val_mmmu_acc,0.28889,
+≥3,13000,mmstar_average,0.3264133242561134,
+≥3,13000,ocrbench_ocrbench_accuracy,0.553,
+≥3,13000,seedbench_seed_all,0.5306837131739855,
+≥3,13000,textvqa_val_exact_match,0.5323599999999999,0.0067627001192260856
+≥3,14000,ai2d_exact_match,0.4381476683937824,0.008930032335354969
+≥3,14000,average,0.4678786302971554,
+≥3,14000,average_rank,2.8,
+≥3,14000,chartqa_relaxed_overall,0.6104,0.009755142291143075
+≥3,14000,docvqa_val_anls,0.6523739582747238,0.006065891171788989
+≥3,14000,infovqa_val_anls,0.2541881734588241,0.006851623469491799
+≥3,14000,mme_total_score,1188.5243097238895,
+≥3,14000,mmmu_val_mmmu_acc,0.30333,
+≥3,14000,mmstar_average,0.3360144984503474,
+≥3,14000,ocrbench_ocrbench_accuracy,0.544,
+≥3,14000,seedbench_seed_all,0.5320733740967204,
+≥3,14000,textvqa_val_exact_match,0.54038,0.006754155375986593
+≥3,15000,ai2d_exact_match,0.4420336787564767,0.008938473522297184
+≥3,15000,average,0.4717225541426424,
+≥3,15000,average_rank,2.6,
+≥3,15000,chartqa_relaxed_overall,0.6204,0.009707689307588963
+≥3,15000,docvqa_val_anls,0.6657062061222615,0.005987913582977679
+≥3,15000,infovqa_val_anls,0.2547770182395889,0.00697535381427897
+≥3,15000,mme_total_score,1150.081532613045,
+≥3,15000,mmmu_val_mmmu_acc,0.29889,
+≥3,15000,mmstar_average,0.32974187627218104,
+≥3,15000,ocrbench_ocrbench_accuracy,0.556,
+≥3,15000,seedbench_seed_all,0.533574207893274,
+≥3,15000,textvqa_val_exact_match,0.54438,0.006740769296908389
+≥3,16000,ai2d_exact_match,0.44397668393782386,0.008942485993062323
+≥3,16000,average,0.4725329505079693,
+≥3,16000,average_rank,2.7,
+≥3,16000,chartqa_relaxed_overall,0.6136,0.009740429476494075
+≥3,16000,docvqa_val_anls,0.6627356411508976,0.00599828184206493
+≥3,16000,infovqa_val_anls,0.25144929243788827,0.006859545868541458
+≥3,16000,mme_total_score,1189.4136654661866,
+≥3,16000,mmmu_val_mmmu_acc,0.30556,
+≥3,16000,mmstar_average,0.32828609880164517,
+≥3,16000,ocrbench_ocrbench_accuracy,0.565,
+≥3,16000,seedbench_seed_all,0.5359088382434686,
+≥3,16000,textvqa_val_exact_match,0.54628,0.006755557699551266
+≥3,17000,ai2d_exact_match,0.4423575129533679,0.008939151893135124
+≥3,17000,average,0.47219284094380815,
+≥3,17000,average_rank,2.9,
+≥3,17000,chartqa_relaxed_overall,0.6196,0.009711645711462604
+≥3,17000,docvqa_val_anls,0.6671354152323413,0.005979986643812461
+≥3,17000,infovqa_val_anls,0.26085018558007095,0.006930202483417548
+≥3,17000,mme_total_score,1181.9268707482993,
+≥3,17000,mmmu_val_mmmu_acc,0.29667,
+≥3,17000,mmstar_average,0.3246494808541184,
+≥3,17000,ocrbench_ocrbench_accuracy,0.556,
+≥3,17000,seedbench_seed_all,0.5353529738743746,
+≥3,17000,textvqa_val_exact_match,0.5471199999999999,0.006741055517194408
+≥3,18000,ai2d_exact_match,0.44624352331606215,0.0089469921763539
+≥3,18000,average,0.47727537354972976,
+≥3,18000,average_rank,2.8,
+≥3,18000,chartqa_relaxed_overall,0.6212,0.009703704898413913
+≥3,18000,docvqa_val_anls,0.6676971859833172,0.005968624246725931
+≥3,18000,infovqa_val_anls,0.2614701461784385,0.006943538426265278
+≥3,18000,mme_total_score,1133.047819127651,
+≥3,18000,mmmu_val_mmmu_acc,0.30444,
+≥3,18000,mmstar_average,0.3242292852357318,
+≥3,18000,ocrbench_ocrbench_accuracy,0.582,
+≥3,18000,seedbench_seed_all,0.5367982212340189,
+≥3,18000,textvqa_val_exact_match,0.5513999999999999,0.006735687188133017
+≥3,19000,ai2d_exact_match,0.4520725388601036,0.008957715852675527
+≥3,19000,average,0.4762675915069992,
+≥3,19000,average_rank,3.0,
+≥3,19000,chartqa_relaxed_overall,0.6216,0.009701702181065136
+≥3,19000,docvqa_val_anls,0.6679273632688325,0.00596194457686321
+≥3,19000,infovqa_val_anls,0.25211534311880446,0.006837669178934141
+≥3,19000,mme_total_score,1168.6077430972389,
+≥3,19000,mmmu_val_mmmu_acc,0.30111,
+≥3,19000,mmstar_average,0.334229548576509,
+≥3,19000,ocrbench_ocrbench_accuracy,0.566,
+≥3,19000,seedbench_seed_all,0.5363535297387437,
+≥3,19000,textvqa_val_exact_match,0.555,0.006737661257130932
+≥3,20000,ai2d_exact_match,0.4566062176165803,0.008965198879336198
+≥3,20000,average,0.4782761612786655,
+≥3,20000,average_rank,2.7,
+≥3,20000,chartqa_relaxed_overall,0.6268,0.009675026948726469
+≥3,20000,docvqa_val_anls,0.6699567897644018,0.005975453790424837
+≥3,20000,infovqa_val_anls,0.2594904076423186,0.006910668664574003
+≥3,20000,mme_total_score,1194.4682873149259,
+≥3,20000,mmmu_val_mmmu_acc,0.30667,
+≥3,20000,mmstar_average,0.3291890626103143,
+≥3,20000,ocrbench_ocrbench_accuracy,0.571,
+≥3,20000,seedbench_seed_all,0.5353529738743746,
+≥3,20000,textvqa_val_exact_match,0.54942,0.0067426571472292
+≥4,1000,ai2d_exact_match,0.266839378238342,0.007960790788435024
+≥4,1000,average,0.28718938224797474,
+≥4,1000,average_rank,2.8,
+≥4,1000,chartqa_relaxed_overall,0.3824,0.009721414421746647
+≥4,1000,docvqa_val_anls,0.3742280929549393,0.005897617626003216
+≥4,1000,infovqa_val_anls,0.18767733564942402,0.006495529242061099
+≥4,1000,mme_total_score,970.0657262905162,
+≥4,1000,mmmu_val_mmmu_acc,0.24667,
+≥4,1000,mmstar_average,0.20409674845299178,
+≥4,1000,ocrbench_ocrbench_accuracy,0.324,
+≥4,1000,seedbench_seed_all,0.2471928849360756,
+≥4,1000,textvqa_val_exact_match,0.3516,0.006519815150594346
+≥4,2000,ai2d_exact_match,0.2700777202072539,0.007991243694641088
+≥4,2000,average,0.32538295993176786,
+≥4,2000,average_rank,2.8,
+≥4,2000,chartqa_relaxed_overall,0.476,0.009990471651004463
+≥4,2000,docvqa_val_anls,0.45055679456484166,0.006087636141467791
+≥4,2000,infovqa_val_anls,0.21184608413063888,0.006740983882332282
+≥4,2000,mme_total_score,1065.059423769508,
+≥4,2000,mmmu_val_mmmu_acc,0.25444,
+≥4,2000,mmstar_average,0.20479630173942967,
+≥4,2000,ocrbench_ocrbench_accuracy,0.404,
+≥4,2000,seedbench_seed_all,0.2535297387437465,
+≥4,2000,textvqa_val_exact_match,0.4032,0.00669032914742019
+≥4,3000,ai2d_exact_match,0.26813471502590674,0.007973037037795191
+≥4,3000,average,0.3429973351943505,
+≥4,3000,average_rank,3.6,
+≥4,3000,chartqa_relaxed_overall,0.5052,0.010001459677380663
+≥4,3000,docvqa_val_anls,0.4883627712637139,0.006123671768321872
+≥4,3000,infovqa_val_anls,0.2020989926298624,0.006492359244043468
+≥4,3000,mme_total_score,1028.0742296918768,
+≥4,3000,mmmu_val_mmmu_acc,0.24444,
+≥4,3000,mmstar_average,0.23755500197641968,
+≥4,3000,ocrbench_ocrbench_accuracy,0.417,
+≥4,3000,seedbench_seed_all,0.2961645358532518,
+≥4,3000,textvqa_val_exact_match,0.42802000000000007,0.006729073636571477
+≥4,4000,ai2d_exact_match,0.297279792746114,0.008226320033454882
+≥4,4000,average,0.37640705986204226,
+≥4,4000,average_rank,3.0,
+≥4,4000,chartqa_relaxed_overall,0.5328,0.009980456292330589
+≥4,4000,docvqa_val_anls,0.5114700599486628,0.006120071866795458
+≥4,4000,infovqa_val_anls,0.20557836945629954,0.006329851460183733
+≥4,4000,mme_total_score,1074.640656262505,
+≥4,4000,mmmu_val_mmmu_acc,0.25889,
+≥4,4000,mmstar_average,0.24011275407256244,
+≥4,4000,ocrbench_ocrbench_accuracy,0.489,
+≥4,4000,seedbench_seed_all,0.40261256253474154,
+≥4,4000,textvqa_val_exact_match,0.44992,0.006773387223162055
+≥4,5000,ai2d_exact_match,0.32998704663212436,0.008462949140760363
+≥4,5000,average,0.3995227518308942,
+≥4,5000,average_rank,2.8,
+≥4,5000,chartqa_relaxed_overall,0.55,0.009951864943131942
+≥4,5000,docvqa_val_anls,0.5627332434349699,0.006167596088104117
+≥4,5000,infovqa_val_anls,0.20676909019266723,0.0063195922615256655
+≥4,5000,mme_total_score,1081.3841536614646,
+≥4,5000,mmmu_val_mmmu_acc,0.26667,
+≥4,5000,mmstar_average,0.26742033896981504,
+≥4,5000,ocrbench_ocrbench_accuracy,0.49,
+≥4,5000,seedbench_seed_all,0.4530850472484714,
+≥4,5000,textvqa_val_exact_match,0.46903999999999996,0.006785801728684695
+≥4,6000,ai2d_exact_match,0.35654145077720206,0.008620788425978479
+≥4,6000,average,0.41458714913417777,
+≥4,6000,average_rank,2.8,
+≥4,6000,chartqa_relaxed_overall,0.5632,0.009921778100334079
+≥4,6000,docvqa_val_anls,0.5818014224607982,0.006182490179642956
+≥4,6000,infovqa_val_anls,0.2145391217079547,0.006472934595237677
+≥4,6000,mme_total_score,1132.2886154461785,
+≥4,6000,mmmu_val_mmmu_acc,0.26667,
+≥4,6000,mmstar_average,0.28714914548287923,
+≥4,6000,ocrbench_ocrbench_accuracy,0.499,
+≥4,6000,seedbench_seed_all,0.47376320177876596,
+≥4,6000,textvqa_val_exact_match,0.48862,0.006787319991169747
+≥4,7000,ai2d_exact_match,0.38471502590673573,0.008756678690415541
+≥4,7000,average,0.42592935170009355,
+≥4,7000,average_rank,3.1,
+≥4,7000,chartqa_relaxed_overall,0.5804,0.009871844677005952
+≥4,7000,docvqa_val_anls,0.5710623718792285,0.006078423874650784
+≥4,7000,infovqa_val_anls,0.22007869704137703,0.006475129444868969
+≥4,7000,mme_total_score,1041.2597038815525,
+≥4,7000,mmmu_val_mmmu_acc,0.28444,
+≥4,7000,mmstar_average,0.3026487819798931,
+≥4,7000,ocrbench_ocrbench_accuracy,0.502,
+≥4,7000,seedbench_seed_all,0.4947192884936076,
+≥4,7000,textvqa_val_exact_match,0.4933,0.006785560460724908
+≥4,8000,ai2d_exact_match,0.3915155440414508,0.008784780895708938
+≥4,8000,average,0.43659006376695136,
+≥4,8000,average_rank,3.0,
+≥4,8000,chartqa_relaxed_overall,0.5736,0.009893046292521752
+≥4,8000,docvqa_val_anls,0.6079864136742988,0.006139878520335163
+≥4,8000,infovqa_val_anls,0.23243402779245617,0.006686893363455147
+≥4,8000,mme_total_score,1108.9173669467787,
+≥4,8000,mmmu_val_mmmu_acc,0.28,
+≥4,8000,mmstar_average,0.3276025817239844,
+≥4,8000,ocrbench_ocrbench_accuracy,0.508,
+≥4,8000,seedbench_seed_all,0.5016120066703724,
+≥4,8000,textvqa_val_exact_match,0.50656,0.006805281452749051
+≥4,9000,ai2d_exact_match,0.39248704663212436,0.008788649010397578
+≥4,9000,average,0.4379212821083599,
+≥4,9000,average_rank,2.9,
+≥4,9000,chartqa_relaxed_overall,0.5844,0.009858475126140203
+≥4,9000,docvqa_val_anls,0.6225000882770518,0.00610983265425905
+≥4,9000,infovqa_val_anls,0.2357319670089269,0.006735352134813103
+≥4,9000,mme_total_score,1054.3165266106444,
+≥4,9000,mmmu_val_mmmu_acc,0.28556,
+≥4,9000,mmstar_average,0.30919474945291153,
+≥4,9000,ocrbench_ocrbench_accuracy,0.496,
+≥4,9000,seedbench_seed_all,0.5078376876042245,
+≥4,9000,textvqa_val_exact_match,0.50758,0.0067866133191798106
+≥4,10000,ai2d_exact_match,0.4177461139896373,0.008876547725654098
+≥4,10000,average,0.4482945169324334,
+≥4,10000,average_rank,3.0,
+≥4,10000,chartqa_relaxed_overall,0.5872,0.009848718845878486
+≥4,10000,docvqa_val_anls,0.6178172719068701,0.006018237392964321
+≥4,10000,infovqa_val_anls,0.24180220451279583,0.006673139519957623
+≥4,10000,mme_total_score,1143.5380152060825,
+≥4,10000,mmmu_val_mmmu_acc,0.29667,
+≥4,10000,mmstar_average,0.31635030378359796,
+≥4,10000,ocrbench_ocrbench_accuracy,0.524,
+≥4,10000,seedbench_seed_all,0.5165647581989995,
+≥4,10000,textvqa_val_exact_match,0.5165,0.006796704277648658
+≥4,11000,ai2d_exact_match,0.41515544041450775,0.00886864516657515
+≥4,11000,average,0.45134109009976725,
+≥4,11000,average_rank,2.5,
+≥4,11000,chartqa_relaxed_overall,0.5956,0.009817474681589429
+≥4,11000,docvqa_val_anls,0.629269001239484,0.00608373788497042
+≥4,11000,infovqa_val_anls,0.24324006994727237,0.006777064540159464
+≥4,11000,mme_total_score,1228.8085234093637,
+≥4,11000,mmmu_val_mmmu_acc,0.28333,
+≥4,11000,mmstar_average,0.3288790569397762,
+≥4,11000,ocrbench_ocrbench_accuracy,0.522,
+≥4,11000,seedbench_seed_all,0.525236242356865,
+≥4,11000,textvqa_val_exact_match,0.5193599999999999,0.0067761804436039675
+≥4,12000,ai2d_exact_match,0.4183937823834197,0.008878484004260249
+≥4,12000,average,0.45687238598965646,
+≥4,12000,average_rank,3.3,
+≥4,12000,chartqa_relaxed_overall,0.5988,0.0098047885010856
+≥4,12000,docvqa_val_anls,0.6281800356608191,0.005956403319187123
+≥4,12000,infovqa_val_anls,0.242249391011484,0.006664412716854741
+≥4,12000,mme_total_score,1051.548619447779,
+≥4,12000,mmmu_val_mmmu_acc,0.28,
+≥4,12000,mmstar_average,0.32638661949265274,
+≥4,12000,ocrbench_ocrbench_accuracy,0.553,
+≥4,12000,seedbench_seed_all,0.5309616453585325,
+≥4,12000,textvqa_val_exact_match,0.53388,0.006762808309810877
+≥4,13000,ai2d_exact_match,0.4323186528497409,0.008916326937351901
+≥4,13000,average,0.46134498058034357,
+≥4,13000,average_rank,3.1,
+≥4,13000,chartqa_relaxed_overall,0.5948,0.009820578470976232
+≥4,13000,docvqa_val_anls,0.6459204882256453,0.006047391420582867
+≥4,13000,infovqa_val_anls,0.24395762124162781,0.006787945348887751
+≥4,13000,mme_total_score,1195.637755102041,
+≥4,13000,mmmu_val_mmmu_acc,0.29556,
+≥4,13000,mmstar_average,0.329867095702076,
+≥4,13000,ocrbench_ocrbench_accuracy,0.542,
+≥4,13000,seedbench_seed_all,0.5337409672040022,
+≥4,13000,textvqa_val_exact_match,0.53394,0.006767804364428913
+≥4,14000,ai2d_exact_match,0.4319948186528497,0.008915528710615492
+≥4,14000,average,0.4668148142530245,
+≥4,14000,average_rank,2.7,
+≥4,14000,chartqa_relaxed_overall,0.6076,0.009767653701044555
+≥4,14000,docvqa_val_anls,0.6561789267585798,0.005953346874132679
+≥4,14000,infovqa_val_anls,0.24945371957223306,0.006769327490532885
+≥4,14000,mme_total_score,1259.298019207683,
+≥4,14000,mmmu_val_mmmu_acc,0.30111,
+≥4,14000,mmstar_average,0.32172026573936097,
+≥4,14000,ocrbench_ocrbench_accuracy,0.55,
+≥4,14000,seedbench_seed_all,0.5360755975541968,
+≥4,14000,textvqa_val_exact_match,0.5472000000000001,0.006748951153204005
+≥4,15000,ai2d_exact_match,0.44624352331606215,0.008946992176353898
+≥4,15000,average,0.46662671868754135,
+≥4,15000,average_rank,2.9,
+≥4,15000,chartqa_relaxed_overall,0.6044,0.009781540134915584
+≥4,15000,docvqa_val_anls,0.6622581274446402,0.005962189435141322
+≥4,15000,infovqa_val_anls,0.2534140745372918,0.006885986461871116
+≥4,15000,mme_total_score,1200.2537014805923,
+≥4,15000,mmmu_val_mmmu_acc,0.29,
+≥4,15000,mmstar_average,0.3198168607331238,
+≥4,15000,ocrbench_ocrbench_accuracy,0.538,
+≥4,15000,seedbench_seed_all,0.5381878821567537,
+≥4,15000,textvqa_val_exact_match,0.54732,0.006746470669416614
+≥4,16000,ai2d_exact_match,0.43911917098445596,0.008932194723472647
+≥4,16000,average,0.46812785270927354,
+≥4,16000,average_rank,3.0,
+≥4,16000,chartqa_relaxed_overall,0.6108,0.00975332737879659
+≥4,16000,docvqa_val_anls,0.6699643513666329,0.005944732124585459
+≥4,16000,infovqa_val_anls,0.2589072217280723,0.006864729360775582
+≥4,16000,mme_total_score,1239.2577030812326,
+≥4,16000,mmmu_val_mmmu_acc,0.28444,
+≥4,16000,mmstar_average,0.3155913699930164,
+≥4,16000,ocrbench_ocrbench_accuracy,0.547,
+≥4,16000,seedbench_seed_all,0.535408560311284,
+≥4,16000,textvqa_val_exact_match,0.55192,0.006727935474062503
+≥4,17000,ai2d_exact_match,0.44591968911917096,0.00894635996642554
+≥4,17000,average,0.4711902865454063,
+≥4,17000,average_rank,2.9,
+≥4,17000,chartqa_relaxed_overall,0.6092,0.009760545645634788
+≥4,17000,docvqa_val_anls,0.6630256300679175,0.005926991608870499
+≥4,17000,infovqa_val_anls,0.2604941623528308,0.0069459226352746855
+≥4,17000,mme_total_score,1231.9475790316128,
+≥4,17000,mmmu_val_mmmu_acc,0.28667,
+≥4,17000,mmstar_average,0.3281863880858027,
+≥4,17000,ocrbench_ocrbench_accuracy,0.559,
+≥4,17000,seedbench_seed_all,0.5380767092829349,
+≥4,17000,textvqa_val_exact_match,0.55014,0.00673464677421427
+≥4,18000,ai2d_exact_match,0.44527202072538863,0.008945084019331404
+≥4,18000,average,0.4730863890198541,
+≥4,18000,average_rank,3.1,
+≥4,18000,chartqa_relaxed_overall,0.6148,0.00973479791861169
+≥4,18000,docvqa_val_anls,0.6724670614264582,0.0059283840951577715
+≥4,18000,infovqa_val_anls,0.2591524677671406,0.006860568910244235
+≥4,18000,mme_total_score,1230.187074829932,
+≥4,18000,mmmu_val_mmmu_acc,0.28222,
+≥4,18000,mmstar_average,0.3313130996754855,
+≥4,18000,ocrbench_ocrbench_accuracy,0.559,
+≥4,18000,seedbench_seed_all,0.5391328515842134,
+≥4,18000,textvqa_val_exact_match,0.55442,0.0067378017419973775
+≥4,19000,ai2d_exact_match,0.4475388601036269,0.008949482610884277
+≥4,19000,average,0.4748981492546839,
+≥4,19000,average_rank,3.0,
+≥4,19000,chartqa_relaxed_overall,0.614,0.009738559226822298
+≥4,19000,docvqa_val_anls,0.6780110114952748,0.005954038851856335
+≥4,19000,infovqa_val_anls,0.2592553130284412,0.0068947091925615645
+≥4,19000,mme_total_score,1280.6934773909566,
+≥4,19000,mmmu_val_mmmu_acc,0.29778,
+≥4,19000,mmstar_average,0.33015385627459537,
+≥4,19000,ocrbench_ocrbench_accuracy,0.55,
+≥4,19000,seedbench_seed_all,0.5397443023902168,
+≥4,19000,textvqa_val_exact_match,0.5576000000000001,0.00671993150976252
+≥4,20000,ai2d_exact_match,0.45077720207253885,0.008955440137395838
+≥4,20000,average,0.47750231781976243,
+≥4,20000,average_rank,2.7,
+≥4,20000,chartqa_relaxed_overall,0.6204,0.009707689307588963
+≥4,20000,docvqa_val_anls,0.673153386104693,0.005945073379634221
+≥4,20000,infovqa_val_anls,0.2604945747241511,0.006917780880359967
+≥4,20000,mme_total_score,1348.6498599439776,
+≥4,20000,mmmu_val_mmmu_acc,0.29222,
+≥4,20000,mmstar_average,0.32618642565880246,
+≥4,20000,ocrbench_ocrbench_accuracy,0.569,
+≥4,20000,seedbench_seed_all,0.5406892718176765,
+≥4,20000,textvqa_val_exact_match,0.5646,0.006722023885782034
+≥5,1000,ai2d_exact_match,0.27396373056994816,0.008027076080717028
+≥5,1000,average,0.26438200891802877,
+≥5,1000,average_rank,3.0,
+≥5,1000,chartqa_relaxed_overall,0.2832,0.00901285729603301
+≥5,1000,docvqa_val_anls,0.32055326545515606,0.0056245129740867565
+≥5,1000,infovqa_val_anls,0.15327397474830004,0.005916826112508726
+≥5,1000,mme_total_score,1087.6624649859943,
+≥5,1000,mmmu_val_mmmu_acc,0.29778,
+≥5,1000,mmstar_average,0.26060215117868224,
+≥5,1000,ocrbench_ocrbench_accuracy,0.259,
+≥5,1000,seedbench_seed_all,0.2649249583101723,
+≥5,1000,textvqa_val_exact_match,0.26614,0.006037548383085275
+≥5,2000,ai2d_exact_match,0.26392487046632124,0.007932917099101329
+≥5,2000,average,0.2929576826335877,
+≥5,2000,average_rank,3.3,
+≥5,2000,chartqa_relaxed_overall,0.3824,0.009721414421746647
+≥5,2000,docvqa_val_anls,0.3929824030686217,0.005977850940256623
+≥5,2000,infovqa_val_anls,0.15895135963621443,0.005878593482981634
+≥5,2000,mme_total_score,1073.4139655862346,
+≥5,2000,mmmu_val_mmmu_acc,0.27333,
+≥5,2000,mmstar_average,0.25335477956948643,
+≥5,2000,ocrbench_ocrbench_accuracy,0.301,
+≥5,2000,seedbench_seed_all,0.26831573096164535,
+≥5,2000,textvqa_val_exact_match,0.34236,0.006479253215027554
+≥5,3000,ai2d_exact_match,0.2600388601036269,0.007895056974601723
+≥5,3000,average,0.3126381365493242,
+≥5,3000,average_rank,3.9,
+≥5,3000,chartqa_relaxed_overall,0.4324,0.009910165515884228
+≥5,3000,docvqa_val_anls,0.4366357263318607,0.00610598785442012
+≥5,3000,infovqa_val_anls,0.17846201123654198,0.006273305639193489
+≥5,3000,mme_total_score,1164.5565226090434,
+≥5,3000,mmmu_val_mmmu_acc,0.28778,
+≥5,3000,mmstar_average,0.25130117268378377,
+≥5,3000,ocrbench_ocrbench_accuracy,0.344,
+≥5,3000,seedbench_seed_all,0.2858254585881045,
+≥5,3000,textvqa_val_exact_match,0.3373,0.006457064405451384
+≥5,4000,ai2d_exact_match,0.25647668393782386,0.007859644922870104
+≥5,4000,average,0.3300809923443584,
+≥5,4000,average_rank,4.3,
+≥5,4000,chartqa_relaxed_overall,0.4428,0.009936335154498413
+≥5,4000,docvqa_val_anls,0.4736486989184438,0.006240863735639683
+≥5,4000,infovqa_val_anls,0.19267658764675277,0.006512420811238904
+≥5,4000,mme_total_score,1218.2668067226891,
+≥5,4000,mmmu_val_mmmu_acc,0.26889,
+≥5,4000,mmstar_average,0.22297093502644408,
+≥5,4000,ocrbench_ocrbench_accuracy,0.379,
+≥5,4000,seedbench_seed_all,0.322846025569761,
+≥5,4000,textvqa_val_exact_match,0.41142,0.006712445761838313
+≥5,5000,ai2d_exact_match,0.25161917098445596,0.007810248924722509
+≥5,5000,average,0.3420574749713038,
+≥5,5000,average_rank,4.2,
+≥5,5000,chartqa_relaxed_overall,0.4488,0.009949423119365426
+≥5,5000,docvqa_val_anls,0.4973120888104521,0.00627054301371889
+≥5,5000,infovqa_val_anls,0.20687122924296383,0.006767419172429617
+≥5,5000,mme_total_score,1285.299119647859,
+≥5,5000,mmmu_val_mmmu_acc,0.26778,
+≥5,5000,mmstar_average,0.24681232878335083,
+≥5,5000,ocrbench_ocrbench_accuracy,0.392,
+≥5,5000,seedbench_seed_all,0.3604224569205114,
+≥5,5000,textvqa_val_exact_match,0.4069,0.00670861230775927
+≥5,6000,ai2d_exact_match,0.2704015544041451,0.00799425923314582
+≥5,6000,average,0.35916516291601697,
+≥5,6000,average_rank,4.4,
+≥5,6000,chartqa_relaxed_overall,0.4844,0.009997131241172205
+≥5,6000,docvqa_val_anls,0.5108154498847224,0.0062636540505031655
+≥5,6000,infovqa_val_anls,0.20262763630072025,0.0066138397079363274
+≥5,6000,mme_total_score,1273.862545018007,
+≥5,6000,mmmu_val_mmmu_acc,0.27444,
+≥5,6000,mmstar_average,0.2588150329919745,
+≥5,6000,ocrbench_ocrbench_accuracy,0.403,
+≥5,6000,seedbench_seed_all,0.4082267926625903,
+≥5,6000,textvqa_val_exact_match,0.41976,0.006731520716318925
+≥5,7000,ai2d_exact_match,0.31994818652849744,0.008395421656067303
+≥5,7000,average,0.3723337802797541,
+≥5,7000,average_rank,4.5,
+≥5,7000,chartqa_relaxed_overall,0.476,0.009990471651004463
+≥5,7000,docvqa_val_anls,0.5291779466505276,0.006267960743408816
+≥5,7000,infovqa_val_anls,0.20957812087727798,0.006721757004150909
+≥5,7000,mme_total_score,1327.2439975990396,
+≥5,7000,mmmu_val_mmmu_acc,0.27222,
+≥5,7000,mmstar_average,0.29354698358099446,
+≥5,7000,ocrbench_ocrbench_accuracy,0.403,
+≥5,7000,seedbench_seed_all,0.42301278488048916,
+≥5,7000,textvqa_val_exact_match,0.42452,0.006734688198055274
+≥5,8000,ai2d_exact_match,0.30958549222797926,0.008321027166750249
+≥5,8000,average,0.37926040717793597,
+≥5,8000,average_rank,4.5,
+≥5,8000,chartqa_relaxed_overall,0.5136,0.009998299975543861
+≥5,8000,docvqa_val_anls,0.5386485557171258,0.006250093887872433
+≥5,8000,infovqa_val_anls,0.21347817313272946,0.006767638253739939
+≥5,8000,mme_total_score,1351.172769107643,
+≥5,8000,mmmu_val_mmmu_acc,0.27667,
+≥5,8000,mmstar_average,0.27077818615838667,
+≥5,8000,ocrbench_ocrbench_accuracy,0.406,
+≥5,8000,seedbench_seed_all,0.4538632573652029,
+≥5,8000,textvqa_val_exact_match,0.43072,0.00674498000523754
+≥5,9000,ai2d_exact_match,0.32642487046632124,0.00843949241376102
+≥5,9000,average,0.3915431470529602,
+≥5,9000,average_rank,4.4,
+≥5,9000,chartqa_relaxed_overall,0.5196,0.009994312908659929
+≥5,9000,docvqa_val_anls,0.5447526718541965,0.006277223186340111
+≥5,9000,infovqa_val_anls,0.22534586447558344,0.006943394394173722
+≥5,9000,mme_total_score,1380.0509203681472,
+≥5,9000,mmmu_val_mmmu_acc,0.28222,
+≥5,9000,mmstar_average,0.2981132324115024,
+≥5,9000,ocrbench_ocrbench_accuracy,0.42,
+≥5,9000,seedbench_seed_all,0.45703168426903834,
+≥5,9000,textvqa_val_exact_match,0.4504,0.0067806462400486975
+≥5,10000,ai2d_exact_match,0.3121761658031088,0.008340079044408505
+≥5,10000,average,0.3945344056050298,
+≥5,10000,average_rank,4.4,
+≥5,10000,chartqa_relaxed_overall,0.524,0.009990471651004463
+≥5,10000,docvqa_val_anls,0.5476477162524015,0.006282119242898783
+≥5,10000,infovqa_val_anls,0.2268357982996008,0.007080273138697436
+≥5,10000,mme_total_score,1385.6108443377352,
+≥5,10000,mmmu_val_mmmu_acc,0.29222,
+≥5,10000,mmstar_average,0.29882846925636025,
+≥5,10000,ocrbench_ocrbench_accuracy,0.43,
+≥5,10000,seedbench_seed_all,0.4627015008337966,
+≥5,10000,textvqa_val_exact_match,0.4564,0.006792248149691337
+≥5,11000,ai2d_exact_match,0.3403497409326425,0.008528080007639036
+≥5,11000,average,0.40311924614292627,
+≥5,11000,average_rank,4.2,
+≥5,11000,chartqa_relaxed_overall,0.5404,0.009969297405349211
+≥5,11000,docvqa_val_anls,0.5698821874786791,0.006251823346664307
+≥5,11000,infovqa_val_anls,0.22660700332356035,0.006919487246988994
+≥5,11000,mme_total_score,1358.4087635054022,
+≥5,11000,mmmu_val_mmmu_acc,0.28778,
+≥5,11000,mmstar_average,0.28965012568597354,
+≥5,11000,ocrbench_ocrbench_accuracy,0.436,
+≥5,11000,seedbench_seed_all,0.4714841578654808,
+≥5,11000,textvqa_val_exact_match,0.46592,0.006784225516827446
+≥5,12000,ai2d_exact_match,0.342940414507772,0.008543648986216495
+≥5,12000,average,0.4128131018529697,
+≥5,12000,average_rank,4.3,
+≥5,12000,chartqa_relaxed_overall,0.5548,0.009941746291659784
+≥5,12000,docvqa_val_anls,0.578981486161722,0.00625708617478689
+≥5,12000,infovqa_val_anls,0.2380032381589791,0.007080943870134072
+≥5,12000,mme_total_score,1390.3039215686274,
+≥5,12000,mmmu_val_mmmu_acc,0.28222,
+≥5,12000,mmstar_average,0.3060607822951693,
+≥5,12000,ocrbench_ocrbench_accuracy,0.472,
+≥5,12000,seedbench_seed_all,0.46559199555308506,
+≥5,12000,textvqa_val_exact_match,0.47472,0.006773519058221244
+≥5,13000,ai2d_exact_match,0.33678756476683935,0.008506208807020252
+≥5,13000,average,0.41416266738683244,
+≥5,13000,average_rank,4.5,
+≥5,13000,chartqa_relaxed_overall,0.5564,0.009938164963872337
+≥5,13000,docvqa_val_anls,0.5882749499950303,0.0062089530468064
+≥5,13000,infovqa_val_anls,0.2250291831460855,0.007008754051627638
+≥5,13000,mme_total_score,1463.7286914765905,
+≥5,13000,mmmu_val_mmmu_acc,0.28222,
+≥5,13000,mmstar_average,0.32070873992428756,
+≥5,13000,ocrbench_ocrbench_accuracy,0.475,
+≥5,13000,seedbench_seed_all,0.4624235686492496,
+≥5,13000,textvqa_val_exact_match,0.48062,0.006792356759039414
+≥5,14000,ai2d_exact_match,0.35103626943005184,0.008590489143063932
+≥5,14000,average,0.4197541703337554,
+≥5,14000,average_rank,4.4,
+≥5,14000,chartqa_relaxed_overall,0.5644,0.00991868984106597
+≥5,14000,docvqa_val_anls,0.5968397354218249,0.006216108072191749
+≥5,14000,infovqa_val_anls,0.23493831065135024,0.00713715521281919
+≥5,14000,mme_total_score,1381.4046618647458,
+≥5,14000,mmmu_val_mmmu_acc,0.28778,
+≥5,14000,mmstar_average,0.3178569306745569,
+≥5,14000,ocrbench_ocrbench_accuracy,0.465,
+≥5,14000,seedbench_seed_all,0.46931628682601445,
+≥5,14000,textvqa_val_exact_match,0.49062,0.0067877549928948315
+≥5,15000,ai2d_exact_match,0.3448834196891192,0.008555140353607656
+≥5,15000,average,0.4156222682362929,
+≥5,15000,average_rank,4.5,
+≥5,15000,chartqa_relaxed_overall,0.5544,0.009942625323290008
+≥5,15000,docvqa_val_anls,0.5981327465682499,0.0062119027314077434
+≥5,15000,infovqa_val_anls,0.2430496253387209,0.007241165402150032
+≥5,15000,mme_total_score,1405.2406962785115,
+≥5,15000,mmmu_val_mmmu_acc,0.27667,
+≥5,15000,mmstar_average,0.30769549523760537,
+≥5,15000,ocrbench_ocrbench_accuracy,0.462,
+≥5,15000,seedbench_seed_all,0.4724291272929405,
+≥5,15000,textvqa_val_exact_match,0.48134,0.006785688616050607
+≥5,16000,ai2d_exact_match,0.3555699481865285,0.008615532040064747
+≥5,16000,average,0.41928937760980056,
+≥5,16000,average_rank,4.6,
+≥5,16000,chartqa_relaxed_overall,0.556,0.00993907007952043
+≥5,16000,docvqa_val_anls,0.5950015990375694,0.006217949166028718
+≥5,16000,infovqa_val_anls,0.2429016453664355,0.007192121794741783
+≥5,16000,mme_total_score,1444.7096838735495,
+≥5,16000,mmmu_val_mmmu_acc,0.27222,
+≥5,16000,mmstar_average,0.29597997743741594,
+≥5,16000,ocrbench_ocrbench_accuracy,0.484,
+≥5,16000,seedbench_seed_all,0.4802112284602557,
+≥5,16000,textvqa_val_exact_match,0.49172000000000005,0.006790781344017229
+≥5,17000,ai2d_exact_match,0.35783678756476683,0.008627736835305362
+≥5,17000,average,0.4243671877798907,
+≥5,17000,average_rank,4.2,
+≥5,17000,chartqa_relaxed_overall,0.566,0.00991448025705367
+≥5,17000,docvqa_val_anls,0.6011648636453683,0.006202633675401635
+≥5,17000,infovqa_val_anls,0.24233190899997978,0.007185211142139982
+≥5,17000,mme_total_score,1383.0262104841936,
+≥5,17000,mmmu_val_mmmu_acc,0.29778,
+≥5,17000,mmstar_average,0.30487588800790116,
+≥5,17000,ocrbench_ocrbench_accuracy,0.48,
+≥5,17000,seedbench_seed_all,0.48343524180100056,
+≥5,17000,textvqa_val_exact_match,0.48588000000000003,0.006793096079908642
+≥5,18000,ai2d_exact_match,0.3484455958549223,0.008575797499263314
+≥5,18000,average,0.4229679292209723,
+≥5,18000,average_rank,4.4,
+≥5,18000,chartqa_relaxed_overall,0.5564,0.009938164963872337
+≥5,18000,docvqa_val_anls,0.6015112951191799,0.006202182626672507
+≥5,18000,infovqa_val_anls,0.2406225801562843,0.007159684093951319
+≥5,18000,mme_total_score,1388.7428971588636,
+≥5,18000,mmmu_val_mmmu_acc,0.29444,
+≥5,18000,mmstar_average,0.3048242431646457,
+≥5,18000,ocrbench_ocrbench_accuracy,0.489,
+≥5,18000,seedbench_seed_all,0.48176764869371874,
+≥5,18000,textvqa_val_exact_match,0.4897,0.006784304485905058
+≥5,19000,ai2d_exact_match,0.3552461139896373,0.00861377131101951
+≥5,19000,average,0.4271521214095191,
+≥5,19000,average_rank,4.4,
+≥5,19000,chartqa_relaxed_overall,0.564,0.009919725822025206
+≥5,19000,docvqa_val_anls,0.6030864459750552,0.006186369284106836
+≥5,19000,infovqa_val_anls,0.24933668460761893,0.007278561320407618
+≥5,19000,mme_total_score,1420.4112645058024,
+≥5,19000,mmmu_val_mmmu_acc,0.28889,
+≥5,19000,mmstar_average,0.3146239893029105,
+≥5,19000,ocrbench_ocrbench_accuracy,0.495,
+≥5,19000,seedbench_seed_all,0.48254585881045026,
+≥5,19000,textvqa_val_exact_match,0.49163999999999997,0.006786164784802775
+≥5,20000,ai2d_exact_match,0.3630181347150259,0.008654846701304475
+≥5,20000,average,0.4295614849660391,
+≥5,20000,average_rank,4.6,
+≥5,20000,chartqa_relaxed_overall,0.566,0.00991448025705367
+≥5,20000,docvqa_val_anls,0.6104819342838989,0.006177273769345363
+≥5,20000,infovqa_val_anls,0.24655527159214874,0.007266841528312276
+≥5,20000,mme_total_score,1413.0101040416166,
+≥5,20000,mmmu_val_mmmu_acc,0.28444,
+≥5,20000,mmstar_average,0.32507494461467334,
+≥5,20000,ocrbench_ocrbench_accuracy,0.499,
+≥5,20000,seedbench_seed_all,0.4775430794886048,
+≥5,20000,textvqa_val_exact_match,0.4939399999999999,0.006784796384004054

data/fsx_seq_write.json

@@ -0,0 +1,402 @@
+{
+  "metadata": {
+    "thread_counts": [
+      1,
+      4,
+      8,
+      16,
+      32,
+      64
+    ],
+    "io_size_labels": [
+      "64K",
+      "128K",
+      "256K",
+      "512K",
+      "1M",
+      "2M",
+      "4M",
+      "8M"
+    ]
+  },
+  "data": {
+    "CPUONLY": {
+      "throughput": [
+        [
+          0.13872,
+          0.215392,
+          0.306149,
+          0.388085,
+          0.402049,
+          0.414225,
+          0.51085,
+          0.536985
+        ],
+        [
+          0.558772,
+          0.877262,
+          1.191367,
+          1.464402,
+          1.577687,
+          1.806215,
+          1.94855,
+          2.044548
+        ],
+        [
+          1.072563,
+          1.663224,
+          2.223145,
+          2.71,
+          2.894774,
+          3.354969,
+          3.577757,
+          3.727307
+        ],
+        [
+          1.977645,
+          2.878005,
+          3.483857,
+          3.923781,
+          4.210632,
+          4.107823,
+          4.147979,
+          3.93202
+        ],
+        [
+          2.788277,
+          3.429985,
+          3.812412,
+          4.055162,
+          4.123704,
+          4.04353,
+          3.91526,
+          3.834913
+        ],
+        [
+          3.098059,
+          3.791825,
+          3.999797,
+          3.884959,
+          3.88633,
+          3.768967,
+          3.696033,
+          3.672264
+        ]
+      ],
+      "iops": [
+        [
+          2272.7860922431173,
+          1764.4896665074543,
+          1253.9870499290996,
+          794.7976954655592,
+          411.6982806135481,
+          212.08341039344933,
+          130.77766031617077,
+          68.73406505594922
+        ],
+        [
+          9154.919914482994,
+          7186.53414364802,
+          4879.838227962797,
+          2999.095053470957,
+          1615.551269334618,
+          924.7820713979427,
+          498.8287795234736,
+          261.702085864898
+        ],
+        [
+          17572.87054286037,
+          13625.133820645293,
+          9105.999983864962,
+          5550.079117205878,
+          2964.2482662387088,
+          1717.7441329970663,
+          915.9056725879472,
+          477.0952973244767
+        ],
+        [
+          32401.728002133626,
+          23576.620940445686,
+          14269.87690918818,
+          8035.904171279055,
+          4311.687216332899,
+          2103.2055418192667,
+          1061.882694749445,
+          503.29856741281907
+        ],
+        [
+          45683.137702529704,
+          28098.44081400338,
+          15615.638082772872,
+          8304.971730492314,
+          4222.672483484196,
+          2070.287298638386,
+          1002.3064933565628,
+          490.8689124619944
+        ],
+        [
+          50758.594542068495,
+          31062.631157518714,
+          16383.17037191755,
+          7956.39646828034,
+          3979.601702963791,
+          1929.7112757319749,
+          946.1845321024515,
+          470.0498535435535
+        ]
+      ]
+    },
+    "CPU_GPU": {
+      "throughput": [
+        [
+          0.127757,
+          0.184055,
+          0.253168,
+          0.303933,
+          0.314527,
+          0.342576,
+          0.355429,
+          0.372795
+        ],
+        [
+          0.489819,
+          0.743009,
+          0.965846,
+          1.134284,
+          1.011822,
+          1.240008,
+          1.291386,
+          1.302075
+        ],
+        [
+          0.926178,
+          1.331791,
+          1.435213,
+          1.395823,
+          1.406029,
+          1.401242,
+          1.337034,
+          1.37707
+        ],
+        [
+          1.51809,
+          1.419795,
+          1.41124,
+          1.412265,
+          1.361527,
+          1.301745,
+          1.29914,
+          1.331017
+        ],
+        [
+          1.244187,
+          1.280894,
+          1.329507,
+          1.242694,
+          1.212715,
+          1.136772,
+          1.190421,
+          1.265224
+        ],
+        [
+          1.234811,
+          1.279372,
+          1.246113,
+          1.203173,
+          1.185391,
+          1.166313,
+          1.272076,
+          1.441775
+        ]
+      ],
+      "iops": [
+        [
+          2093.176872177107,
+          1507.7761458185525,
+          1036.9746265271071,
+          622.4549992592137,
+          322.07590038936985,
+          175.39907068733754,
+          90.9898788924712,
+          47.71773657423524
+        ],
+        [
+          8025.190053630053,
+          6086.729376897175,
+          3956.1062308264145,
+          2323.012914032528,
+          1036.1054278362203,
+          634.8842226573488,
+          330.59473644863556,
+          166.66553734827153
+        ],
+        [
+          15174.501399016752,
+          10910.035396570416,
+          5878.631378882711,
+          2858.6450904199623,
+          1439.773366142635,
+          717.4360546721377,
+          342.2806463421564,
+          176.2648984498799
+        ],
+        [
+          24872.38014787567,
+          11630.960067132395,
+          5780.439335314154,
+          2892.3182248680237,
+          1394.203535726493,
+          666.4931998767951,
+          332.5798907171207,
+          170.37013112848007
+        ],
+        [
+          20384.75790846492,
+          10493.083849140936,
+          5445.658888259013,
+          2545.0364923195402,
+          1241.8200114087206,
+          582.0270452268836,
+          304.74777356851035,
+          161.9486939837578
+        ],
+        [
+          20231.14510130481,
+          10480.618372005543,
+          5104.07984186592,
+          2464.0990583552143,
+          1213.8407285720978,
+          597.152222653312,
+          325.6515756855089,
+          184.54715088795263
+        ]
+      ]
+    },
+    "GPUD": {
+      "throughput": [
+        [
+          0.135424,
+          0.213732,
+          0.287542,
+          0.365955,
+          0.381534,
+          0.437018,
+          0.472723,
+          0.48824
+        ],
+        [
+          0.531047,
+          0.844527,
+          1.138279,
+          1.408723,
+          1.499822,
+          1.681777,
+          1.820272,
+          1.896262
+        ],
+        [
+          1.022239,
+          1.593599,
+          2.124232,
+          2.583386,
+          2.766715,
+          3.129373,
+          3.360252,
+          3.518104
+        ],
+        [
+          1.849517,
+          2.70284,
+          3.374121,
+          3.876813,
+          3.965969,
+          3.672806,
+          3.918975,
+          3.888437
+        ],
+        [
+          2.773395,
+          3.471455,
+          3.752517,
+          4.028096,
+          3.924482,
+          3.765881,
+          3.707045,
+          3.901955
+        ],
+        [
+          3.112637,
+          3.498519,
+          3.632984,
+          3.727544,
+          3.719021,
+          3.569274,
+          3.751826,
+          3.802593
+        ]
+      ],
+      "iops": [
+        [
+          2218.784374393393,
+          1750.894255562167,
+          1177.7726775350632,
+          749.4757927835702,
+          390.6906854712861,
+          223.75325885880522,
+          121.01697197969126,
+          62.494678019794925
+        ],
+        [
+          8700.672141705247,
+          6918.368324543275,
+          4662.3904932857795,
+          2885.0655830707506,
+          1535.817839540117,
+          861.069790982218,
+          465.9896625478722,
+          242.7214889354055
+        ],
+        [
+          16748.37064961359,
+          13054.763734630178,
+          8700.855317460486,
+          5290.774724246565,
+          2833.1157831595046,
+          1602.2389817201313,
+          860.224446673275,
+          450.31737305657634
+        ],
+        [
+          30302.482789757647,
+          22141.66829462241,
+          13820.397667459622,
+          7939.713667018539,
+          4061.151934854281,
+          1880.476824114228,
+          1003.2576429966564,
+          497.71994631057765
+        ],
+        [
+          45439.30327651579,
+          28438.163353779502,
+          15370.310188652174,
+          8249.540190922035,
+          4018.6694004701576,
+          1928.1312423850352,
+          949.0035300987968,
+          499.4502850206581
+        ],
+        [
+          50997.45005238404,
+          28659.87090428627,
+          14880.70076305479,
+          7634.009662314824,
+          3808.2775805738934,
+          1827.468091883358,
+          960.4675337218762,
+          486.7318762577807
+        ]
+      ]
+    }
+  }
+}

data/lab_allpass_comparison.json

@@ -0,0 +1,147 @@
+{
+  "closed": [
+    {
+      "model": "Fable 5",
+      "all_pass": 0.133,
+      "cost_usd": 1823.1,
+      "input_price_per_mtok": 10.0,
+      "output_price_per_mtok": 50.0,
+      "pricing_source": "Anthropic",
+      "long_context_runs": 0
+    },
+    {
+      "model": "Opus 4.7",
+      "all_pass": 0.071,
+      "cost_usd": 911.6,
+      "input_price_per_mtok": 5.0,
+      "output_price_per_mtok": 25.0,
+      "pricing_source": "Anthropic",
+      "long_context_runs": 0
+    },
+    {
+      "model": "Sonnet 4.6",
+      "all_pass": 0.054,
+      "cost_usd": 546.9,
+      "input_price_per_mtok": 3.0,
+      "output_price_per_mtok": 15.0,
+      "pricing_source": "Anthropic",
+      "long_context_runs": 0
+    },
+    {
+      "model": "Opus 4.6",
+      "all_pass": 0.042,
+      "cost_usd": 911.6,
+      "input_price_per_mtok": 5.0,
+      "output_price_per_mtok": 25.0,
+      "pricing_source": "Anthropic",
+      "long_context_runs": 0
+    },
+    {
+      "model": "GPT-5.5",
+      "all_pass": 0.021,
+      "cost_usd": 919.1,
+      "input_price_per_mtok": 5.0,
+      "output_price_per_mtok": 30.0,
+      "pricing_source": "OpenAI",
+      "long_context_runs": 0
+    },
+    {
+      "model": "Gemini 3.5 Flash",
+      "all_pass": 0.008,
+      "cost_usd": 275.7,
+      "input_price_per_mtok": 1.5,
+      "output_price_per_mtok": 9.0,
+      "pricing_source": "Google",
+      "long_context_runs": 0
+    },
+    {
+      "model": "GPT-5.4",
+      "all_pass": 0.004,
+      "cost_usd": 459.5,
+      "input_price_per_mtok": 2.5,
+      "output_price_per_mtok": 15.0,
+      "pricing_source": "OpenAI",
+      "long_context_runs": 0
+    },
+    {
+      "model": "Gemini 3.1 Pro",
+      "all_pass": 0.0,
+      "cost_usd": 367.6,
+      "input_price_per_mtok": 2.0,
+      "output_price_per_mtok": 12.0,
+      "pricing_source": "Google",
+      "long_context_runs": 0
+    },
+    {
+      "model": "GPT-5.4 mini",
+      "all_pass": 0.0,
+      "cost_usd": 137.9,
+      "input_price_per_mtok": 0.75,
+      "output_price_per_mtok": 4.5,
+      "pricing_source": "OpenAI",
+      "long_context_runs": 0
+    },
+    {
+      "model": "Gemini 3 Flash",
+      "all_pass": 0.0,
+      "cost_usd": 91.9,
+      "input_price_per_mtok": 0.5,
+      "output_price_per_mtok": 3.0,
+      "pricing_source": "Google",
+      "long_context_runs": 0
+    },
+    {
+      "model": "Gemini 3.1 Flash-Lite",
+      "all_pass": 0.0,
+      "cost_usd": 46.0,
+      "input_price_per_mtok": 0.25,
+      "output_price_per_mtok": 1.5,
+      "pricing_source": "Google",
+      "long_context_runs": 0
+    }
+  ],
+  "ours": [
+    {
+      "model": "DeepSeek V4 Pro",
+      "baseline_all_pass": 0.0,
+      "optimized_all_pass": 0.05,
+      "harness": "full frontier",
+      "cost_usd": 77.3,
+      "input_price_per_mtok": 0.435,
+      "output_price_per_mtok": 0.87,
+      "pricing_source": "DeepSeek",
+      "long_context_runs": 0
+    },
+    {
+      "model": "Nemotron-3 Ultra 550B",
+      "baseline_all_pass": 0.0,
+      "optimized_all_pass": 0.03,
+      "harness": "landing gate",
+      "cost_usd": 90.7,
+      "input_price_per_mtok": 0.5,
+      "output_price_per_mtok": 2.2,
+      "pricing_source": "OpenRouter",
+      "long_context_runs": 0
+    },
+    {
+      "model": "DeepSeek V4 Flash",
+      "baseline_all_pass": 0.0,
+      "optimized_all_pass": 0.02,
+      "harness": "landing gate",
+      "cost_usd": 24.9,
+      "input_price_per_mtok": 0.14,
+      "output_price_per_mtok": 0.28,
+      "pricing_source": "DeepSeek",
+      "long_context_runs": 0
+    }
+  ],
+  "n_test": 100,
+  "cost_assumption": {
+    "kind": "agent_rollout_only",
+    "token_trace": "DeepSeek V4 Pro baseline on the 100-task test split",
+    "runs": 100,
+    "input_tokens": 174799637,
+    "output_tokens": 1502524,
+    "limitations": "Uses the same input/output token counts for every model; excludes the common Sonnet judge cost, model-specific verbosity/tokenization differences, caching, and long-context surcharges."
+  }
+}

data/lab_baseline_tasks.json

@@ -0,0 +1,1373 @@
+{
+  "harness": "baseline_vanilla",
+  "splits": {
+    "dev": [
+      {
+        "task": "environmental-esg/draft-markup-of-administrative-settlement-agreement",
+        "title": "Draft markup of administrative settlement agreement",
+        "area": "environmental-esg",
+        "work_type": "review",
+        "n_criteria": 53,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "litigation-dispute-resolution/draft-motion-for-summary-judgment",
+        "title": "Draft motion for summary judgment",
+        "area": "litigation-dispute-resolution",
+        "work_type": "draft",
+        "n_criteria": 69,
+        "n_docs": 17,
+        "n_deliverables": 2,
+        "pooled": 0.058,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "energy-natural-resources/draft-power-purchase-agreement",
+        "title": "Draft power purchase agreement",
+        "area": "energy-natural-resources",
+        "work_type": "draft",
+        "n_criteria": 91,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.4579,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "emerging-companies-venture-capital/identify-issues-in-investors-rights-agreement",
+        "title": "Identify issues in investors rights agreement",
+        "area": "emerging-companies-venture-capital",
+        "work_type": "review",
+        "n_criteria": 37,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.4595,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "tax/analyze-section-382-analysis",
+        "title": "Analyze section 382 analysis",
+        "area": "tax",
+        "work_type": "analyze",
+        "n_criteria": 107,
+        "n_docs": 11,
+        "n_deliverables": 2,
+        "pooled": 0.4642,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "banking-finance/identify-term-sheet-issues",
+        "title": "Identify term sheet issues",
+        "area": "banking-finance",
+        "work_type": "review",
+        "n_criteria": 34,
+        "n_docs": 4,
+        "n_deliverables": 1,
+        "pooled": 0.4706,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "insurance/draft-change-of-control-application",
+        "title": "Draft change of control application",
+        "area": "insurance",
+        "work_type": "draft",
+        "n_criteria": 72,
+        "n_docs": 10,
+        "n_deliverables": 2,
+        "pooled": 0.4769,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "arbitration-international-dispute-resolution/analyze-arbitration-award-for-new-york-convention-enforcement-defenses",
+        "title": "Analyze arbitration award for new york convention enforcement defenses",
+        "area": "arbitration-international-dispute-resolution",
+        "work_type": "analyze",
+        "n_criteria": 43,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.5349,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/draft-restrictive-covenant-agreement",
+        "title": "Draft restrictive covenant agreement",
+        "area": "corporate-ma",
+        "work_type": "draft",
+        "n_criteria": 62,
+        "n_docs": 5,
+        "n_deliverables": 2,
+        "pooled": 0.5376,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-governance/research-regulatory-approval-requirements-for-new-fintech-lending-business-line",
+        "title": "Research regulatory approval requirements for new fintech lending business line",
+        "area": "corporate-governance",
+        "work_type": "research",
+        "n_criteria": 52,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.5513,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "funds-asset-management/draft-compliance-manual",
+        "title": "Draft compliance manual",
+        "area": "funds-asset-management",
+        "work_type": "draft",
+        "n_criteria": 101,
+        "n_docs": 10,
+        "n_deliverables": 1,
+        "pooled": 0.6106,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "data-privacy-cybersecurity/identify-issues-in-transfer-impact-assessment",
+        "title": "Identify issues in transfer impact assessment",
+        "area": "data-privacy-cybersecurity",
+        "work_type": "review",
+        "n_criteria": 33,
+        "n_docs": 8,
+        "n_deliverables": 1,
+        "pooled": 0.697,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "trusts-estates-private-client/identify-issues-in-counterpartys-motion-brief",
+        "title": "Identify issues in counterpartys motion brief",
+        "area": "trusts-estates-private-client",
+        "work_type": "analyze",
+        "n_criteria": 35,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.7048,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "bankruptcy-restructuring/extract-loan-agreement-terms/scenario-02",
+        "title": "Extract loan agreement terms / scenario 02",
+        "area": "bankruptcy-restructuring",
+        "work_type": "analyze",
+        "n_criteria": 102,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.7451,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "immigration/identify-compliance-issues-in-employee-i",
+        "title": "Identify compliance issues in employee i",
+        "area": "immigration",
+        "work_type": "review",
+        "n_criteria": 39,
+        "n_docs": 15,
+        "n_deliverables": 1,
+        "pooled": 0.7521,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "structured-finance-securitization/compare-offering-memorandum-against-indenture",
+        "title": "Compare offering memorandum against indenture",
+        "area": "structured-finance-securitization",
+        "work_type": "review",
+        "n_criteria": 37,
+        "n_docs": 3,
+        "n_deliverables": 1,
+        "pooled": 0.7838,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "employment-labor/compare-non",
+        "title": "Compare non",
+        "area": "employment-labor",
+        "work_type": "analyze",
+        "n_criteria": 38,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.7982,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/research-trade-secret-protections-for-departing-employee",
+        "title": "Research trade secret protections for departing employee",
+        "area": "intellectual-property",
+        "work_type": "research",
+        "n_criteria": 62,
+        "n_docs": 9,
+        "n_deliverables": 1,
+        "pooled": 0.8118,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "international-trade-sanctions/compare-entity-details-against-export-control-classification-lists",
+        "title": "Compare entity details against export control classification lists",
+        "area": "international-trade-sanctions",
+        "work_type": "analyze",
+        "n_criteria": 47,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.8156,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "white-collar-defense-investigations/draft-markup-of-deferred-prosecution-agreement",
+        "title": "Draft markup of deferred prosecution agreement",
+        "area": "white-collar-defense-investigations",
+        "work_type": "review",
+        "n_criteria": 52,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.8397,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "antitrust-competition/draft-pre-notification-briefing-paper",
+        "title": "Draft pre notification briefing paper",
+        "area": "antitrust-competition",
+        "work_type": "draft",
+        "n_criteria": 60,
+        "n_docs": 8,
+        "n_deliverables": 2,
+        "pooled": 0.8444,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "healthcare-life-sciences/draft-management-services-agreement",
+        "title": "Draft management services agreement",
+        "area": "healthcare-life-sciences",
+        "work_type": "draft",
+        "n_criteria": 94,
+        "n_docs": 6,
+        "n_deliverables": 2,
+        "pooled": 0.8865,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "real-estate/draft-purchase-and-sale-agreement",
+        "title": "Draft purchase and sale agreement",
+        "area": "real-estate",
+        "work_type": "draft",
+        "n_criteria": 95,
+        "n_docs": 8,
+        "n_deliverables": 2,
+        "pooled": 0.9088,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "capital-markets/analyze-counterparty-markup-of-underwriting-agreement",
+        "title": "Analyze counterparty markup of underwriting agreement",
+        "area": "capital-markets",
+        "work_type": "analyze",
+        "n_criteria": 58,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.9483,
+        "all_pass_rate": 0.0
+      }
+    ],
+    "test": [
+      {
+        "task": "capital-markets/identify-dd-request-list-issues/scenario-01",
+        "title": "Identify dd request list issues / scenario 01",
+        "area": "capital-markets",
+        "work_type": "review",
+        "n_criteria": 35,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-governance/draft-batch-nda-generation",
+        "title": "Draft batch nda generation",
+        "area": "corporate-governance",
+        "work_type": "draft",
+        "n_criteria": 97,
+        "n_docs": 6,
+        "n_deliverables": 11,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/build-deal-points-library/scenario-02",
+        "title": "Build deal points library / scenario 02",
+        "area": "corporate-ma",
+        "work_type": "analyze",
+        "n_criteria": 191,
+        "n_docs": 9,
+        "n_deliverables": 2,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/draft-acquisition-due-diligence",
+        "title": "Draft acquisition due diligence",
+        "area": "corporate-ma",
+        "work_type": "draft",
+        "n_criteria": 64,
+        "n_docs": 31,
+        "n_deliverables": 1,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "emerging-companies-venture-capital/draft-markup-of-series-b-preferred-stock-purchase-agreement",
+        "title": "Draft markup of series b preferred stock purchase agreement",
+        "area": "emerging-companies-venture-capital",
+        "work_type": "review",
+        "n_criteria": 70,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "funds-asset-management/identify-issues-in-investment-advisory-agreement",
+        "title": "Identify issues in investment advisory agreement",
+        "area": "funds-asset-management",
+        "work_type": "review",
+        "n_criteria": 36,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "funds-asset-management/respond-to-comment-memo",
+        "title": "Respond to comment memo",
+        "area": "funds-asset-management",
+        "work_type": "draft",
+        "n_criteria": 82,
+        "n_docs": 10,
+        "n_deliverables": 1,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/draft-statement-of-work-for-enterprise-cloud-migration-services",
+        "title": "Draft statement of work for enterprise cloud migration services",
+        "area": "intellectual-property",
+        "work_type": "draft",
+        "n_criteria": 73,
+        "n_docs": 6,
+        "n_deliverables": 2,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "structured-finance-securitization/identify-issues-in-term-sheet",
+        "title": "Identify issues in term sheet",
+        "area": "structured-finance-securitization",
+        "work_type": "review",
+        "n_criteria": 33,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "structured-finance-securitization/review-securitization-indenture/scenario-02",
+        "title": "Review securitization indenture / scenario 02",
+        "area": "structured-finance-securitization",
+        "work_type": "review",
+        "n_criteria": 39,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.0,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "environmental-esg/draft-permit-application-narrative/scenario-02",
+        "title": "Draft permit application narrative / scenario 02",
+        "area": "environmental-esg",
+        "work_type": "draft",
+        "n_criteria": 71,
+        "n_docs": 8,
+        "n_deliverables": 2,
+        "pooled": 0.0141,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "tax/draft-transfer-pricing-documentation",
+        "title": "Draft transfer pricing documentation",
+        "area": "tax",
+        "work_type": "draft",
+        "n_criteria": 77,
+        "n_docs": 25,
+        "n_deliverables": 1,
+        "pooled": 0.026,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/draft-markup-of-transition-services-agreement-tsa",
+        "title": "Draft markup of transition services agreement tsa",
+        "area": "corporate-ma",
+        "work_type": "review",
+        "n_criteria": 56,
+        "n_docs": 4,
+        "n_deliverables": 2,
+        "pooled": 0.0536,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "bankruptcy-restructuring/draft-voluntary-chapter-11-petition-and-schedules",
+        "title": "Draft voluntary chapter 11 petition and schedules",
+        "area": "bankruptcy-restructuring",
+        "work_type": "draft",
+        "n_criteria": 85,
+        "n_docs": 9,
+        "n_deliverables": 8,
+        "pooled": 0.1412,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "bankruptcy-restructuring/draft-first-day-motions",
+        "title": "Draft first day motions",
+        "area": "bankruptcy-restructuring",
+        "work_type": "draft",
+        "n_criteria": 97,
+        "n_docs": 9,
+        "n_deliverables": 7,
+        "pooled": 0.2474,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-governance/assess-impact-of-ftc-noncompete-ban-on-existing-employment-agreements",
+        "title": "Assess impact of ftc noncompete ban on existing employment agreements",
+        "area": "corporate-governance",
+        "work_type": "analyze",
+        "n_criteria": 57,
+        "n_docs": 22,
+        "n_deliverables": 1,
+        "pooled": 0.2632,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "arbitration-international-dispute-resolution/draft-markup-of-procedural-order",
+        "title": "Draft markup of procedural order",
+        "area": "arbitration-international-dispute-resolution",
+        "work_type": "review",
+        "n_criteria": 55,
+        "n_docs": 6,
+        "n_deliverables": 2,
+        "pooled": 0.2727,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-governance/compare-entity-compliance-checklist-against-annual-requirements",
+        "title": "Compare entity compliance checklist against annual requirements",
+        "area": "corporate-governance",
+        "work_type": "analyze",
+        "n_criteria": 44,
+        "n_docs": 14,
+        "n_deliverables": 1,
+        "pooled": 0.3636,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "immigration/identify-issues-in-consular-processing-packet",
+        "title": "Identify issues in consular processing packet",
+        "area": "immigration",
+        "work_type": "review",
+        "n_criteria": 33,
+        "n_docs": 10,
+        "n_deliverables": 1,
+        "pooled": 0.3939,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-governance/compare-proxy-statement-disclosures-against-sec-regulatory-requirements",
+        "title": "Compare proxy statement disclosures against sec regulatory requirements",
+        "area": "corporate-governance",
+        "work_type": "analyze",
+        "n_criteria": 44,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.5,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "immigration/extract-eligibility-requirements-from-visa-category-regulations",
+        "title": "Extract eligibility requirements from visa category regulations",
+        "area": "immigration",
+        "work_type": "research",
+        "n_criteria": 48,
+        "n_docs": 6,
+        "n_deliverables": 2,
+        "pooled": 0.5208,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "trusts-estates-private-client/compare-charitable-trust-structures-against-client-goals",
+        "title": "Compare charitable trust structures against client goals",
+        "area": "trusts-estates-private-client",
+        "work_type": "analyze",
+        "n_criteria": 55,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.5273,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/draft-asset-purchase-agreement/scenario-01",
+        "title": "Draft asset purchase agreement / scenario 01",
+        "area": "corporate-ma",
+        "work_type": "draft",
+        "n_criteria": 161,
+        "n_docs": 11,
+        "n_deliverables": 6,
+        "pooled": 0.5466,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "litigation-dispute-resolution/review-privilege-log-clawback-review",
+        "title": "Review privilege log clawback review",
+        "area": "litigation-dispute-resolution",
+        "work_type": "review",
+        "n_criteria": 82,
+        "n_docs": 55,
+        "n_deliverables": 2,
+        "pooled": 0.561,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/identify-issues-in-counterparty-discovery-requests",
+        "title": "Identify issues in counterparty discovery requests",
+        "area": "intellectual-property",
+        "work_type": "analyze",
+        "n_criteria": 46,
+        "n_docs": 8,
+        "n_deliverables": 1,
+        "pooled": 0.587,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "international-trade-sanctions/extract-tariff-classifications-from-product-documentation",
+        "title": "Extract tariff classifications from product documentation",
+        "area": "international-trade-sanctions",
+        "work_type": "analyze",
+        "n_criteria": 72,
+        "n_docs": 12,
+        "n_deliverables": 1,
+        "pooled": 0.5972,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "antitrust-competition/draft-antitrust-compliance-training-materials",
+        "title": "Draft antitrust compliance training materials",
+        "area": "antitrust-competition",
+        "work_type": "draft",
+        "n_criteria": 60,
+        "n_docs": 10,
+        "n_deliverables": 2,
+        "pooled": 0.6,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "data-privacy-cybersecurity/identify-issues-in-incident-response-plan",
+        "title": "Identify issues in incident response plan",
+        "area": "data-privacy-cybersecurity",
+        "work_type": "review",
+        "n_criteria": 38,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.6053,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "arbitration-international-dispute-resolution/identify-deficiencies-and-vulnerabilities-in-counterpartys-statement-of-claim",
+        "title": "Identify deficiencies and vulnerabilities in counterpartys statement of claim",
+        "area": "arbitration-international-dispute-resolution",
+        "work_type": "analyze",
+        "n_criteria": 46,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.6087,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/extract-open-source-license-obligations-from-targets-disclosure-schedule",
+        "title": "Extract open source license obligations from targets disclosure schedule",
+        "area": "intellectual-property",
+        "work_type": "analyze",
+        "n_criteria": 55,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.6182,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "banking-finance/draft-credit-compliance-cert",
+        "title": "Draft credit compliance cert",
+        "area": "banking-finance",
+        "work_type": "draft",
+        "n_criteria": 87,
+        "n_docs": 7,
+        "n_deliverables": 3,
+        "pooled": 0.6437,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "trusts-estates-private-client/analyze-counterparty-markup-of-postnuptial-agreement",
+        "title": "Analyze counterparty markup of postnuptial agreement",
+        "area": "trusts-estates-private-client",
+        "work_type": "analyze",
+        "n_criteria": 46,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.6522,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "funds-asset-management/identify-regulatory-and-operational-deficiencies-in-fund-manager-compliance-manual",
+        "title": "Identify regulatory and operational deficiencies in fund manager compliance manual",
+        "area": "funds-asset-management",
+        "work_type": "review",
+        "n_criteria": 49,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.6531,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "bankruptcy-restructuring/draft-restructuring-support-agreement",
+        "title": "Draft restructuring support agreement",
+        "area": "bankruptcy-restructuring",
+        "work_type": "draft",
+        "n_criteria": 64,
+        "n_docs": 8,
+        "n_deliverables": 1,
+        "pooled": 0.6562,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "litigation-dispute-resolution/categorize-document-production-set-by-relevance-and-privilege",
+        "title": "Categorize document production set by relevance and privilege",
+        "area": "litigation-dispute-resolution",
+        "work_type": "analyze",
+        "n_criteria": 67,
+        "n_docs": 25,
+        "n_deliverables": 2,
+        "pooled": 0.6567,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "banking-finance/compare-closing-documents-against-restructuring-conditions",
+        "title": "Compare closing documents against restructuring conditions",
+        "area": "banking-finance",
+        "work_type": "review",
+        "n_criteria": 38,
+        "n_docs": 9,
+        "n_deliverables": 1,
+        "pooled": 0.6579,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "insurance/draft-markup-of-quota-share-reinsurance-treaty",
+        "title": "Draft markup of quota share reinsurance treaty",
+        "area": "insurance",
+        "work_type": "draft",
+        "n_criteria": 65,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.6615,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "funds-asset-management/draft-fund-ppm",
+        "title": "Draft fund ppm",
+        "area": "funds-asset-management",
+        "work_type": "draft",
+        "n_criteria": 194,
+        "n_docs": 9,
+        "n_deliverables": 1,
+        "pooled": 0.6701,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "international-trade-sanctions/identify-issues-in-export-license-application",
+        "title": "Identify issues in export license application",
+        "area": "international-trade-sanctions",
+        "work_type": "analyze",
+        "n_criteria": 46,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.6739,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "energy-natural-resources/compare-power-purchase-agreement-against-term-sheet",
+        "title": "Compare power purchase agreement against term sheet",
+        "area": "energy-natural-resources",
+        "work_type": "review",
+        "n_criteria": 43,
+        "n_docs": 4,
+        "n_deliverables": 1,
+        "pooled": 0.6744,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/review-patent-complaint",
+        "title": "Review patent complaint",
+        "area": "intellectual-property",
+        "work_type": "analyze",
+        "n_criteria": 34,
+        "n_docs": 10,
+        "n_deliverables": 1,
+        "pooled": 0.6765,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "employment-labor/research-wage-and-hour-classification-for-new-job-role",
+        "title": "Research wage and hour classification for new job role",
+        "area": "employment-labor",
+        "work_type": "research",
+        "n_criteria": 41,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.6829,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "bankruptcy-restructuring/draft-first-day-motions-package",
+        "title": "Draft first day motions package",
+        "area": "bankruptcy-restructuring",
+        "work_type": "draft",
+        "n_criteria": 70,
+        "n_docs": 10,
+        "n_deliverables": 6,
+        "pooled": 0.7,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "emerging-companies-venture-capital/analyze-counterparty-markup-of-stock-purchase-agreement",
+        "title": "Analyze counterparty markup of stock purchase agreement",
+        "area": "emerging-companies-venture-capital",
+        "work_type": "analyze",
+        "n_criteria": 47,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.7021,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "capital-markets/extract-key-terms-from-underwriting-agreement",
+        "title": "Extract key terms from underwriting agreement",
+        "area": "capital-markets",
+        "work_type": "analyze",
+        "n_criteria": 64,
+        "n_docs": 4,
+        "n_deliverables": 1,
+        "pooled": 0.7031,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/review-corporate-consents-diligence/scenario-02",
+        "title": "Review corporate consents diligence / scenario 02",
+        "area": "corporate-ma",
+        "work_type": "analyze",
+        "n_criteria": 54,
+        "n_docs": 14,
+        "n_deliverables": 1,
+        "pooled": 0.7037,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/update-data-processing-agreement-template-for-new-gdpr-adequacy-decision",
+        "title": "Update data processing agreement template for new gdpr adequacy decision",
+        "area": "intellectual-property",
+        "work_type": "draft",
+        "n_criteria": 52,
+        "n_docs": 6,
+        "n_deliverables": 2,
+        "pooled": 0.7115,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/analyze-counterparty-markup-of-ip-assignment-agreement",
+        "title": "Analyze counterparty markup of ip assignment agreement",
+        "area": "intellectual-property",
+        "work_type": "analyze",
+        "n_criteria": 59,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.7288,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/draft-open-source-compliance-policy",
+        "title": "Draft open source compliance policy",
+        "area": "intellectual-property",
+        "work_type": "draft",
+        "n_criteria": 51,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.7451,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-governance/draft-action-by-incorporator",
+        "title": "Draft action by incorporator",
+        "area": "corporate-governance",
+        "work_type": "draft",
+        "n_criteria": 45,
+        "n_docs": 4,
+        "n_deliverables": 1,
+        "pooled": 0.7556,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "tax/identify-issues-in-notice-of-deficiency",
+        "title": "Identify issues in notice of deficiency",
+        "area": "tax",
+        "work_type": "analyze",
+        "n_criteria": 41,
+        "n_docs": 8,
+        "n_deliverables": 1,
+        "pooled": 0.7561,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/identify-issues-in-portfolio-company-contracts",
+        "title": "Identify issues in portfolio company contracts",
+        "area": "corporate-ma",
+        "work_type": "analyze",
+        "n_criteria": 58,
+        "n_docs": 8,
+        "n_deliverables": 1,
+        "pooled": 0.7586,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "white-collar-defense-investigations/identify-issues-in-proposed-consent-decree",
+        "title": "Identify issues in proposed consent decree",
+        "area": "white-collar-defense-investigations",
+        "work_type": "review",
+        "n_criteria": 39,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.7692,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/build-deal-points-library/scenario-01",
+        "title": "Build deal points library / scenario 01",
+        "area": "corporate-ma",
+        "work_type": "analyze",
+        "n_criteria": 194,
+        "n_docs": 9,
+        "n_deliverables": 2,
+        "pooled": 0.7784,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "real-estate/analyze-counterparty-markup-of-purchase-and-sale-agreement",
+        "title": "Analyze counterparty markup of purchase and sale agreement",
+        "area": "real-estate",
+        "work_type": "analyze",
+        "n_criteria": 46,
+        "n_docs": 4,
+        "n_deliverables": 1,
+        "pooled": 0.7826,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-governance/review-nda-playbook-review",
+        "title": "Review nda playbook review",
+        "area": "corporate-governance",
+        "work_type": "review",
+        "n_criteria": 57,
+        "n_docs": 3,
+        "n_deliverables": 1,
+        "pooled": 0.7895,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "employment-labor/compare-separation-agreement-against-compensation-survey",
+        "title": "Compare separation agreement against compensation survey",
+        "area": "employment-labor",
+        "work_type": "analyze",
+        "n_criteria": 58,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.7931,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/analyze-responses-to-ddrl/scenario-02",
+        "title": "Analyze responses to ddrl / scenario 02",
+        "area": "corporate-ma",
+        "work_type": "analyze",
+        "n_criteria": 44,
+        "n_docs": 12,
+        "n_deliverables": 1,
+        "pooled": 0.7955,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "litigation-dispute-resolution/extract-key-admissions-from-deposition-transcript",
+        "title": "Extract key admissions from deposition transcript",
+        "area": "litigation-dispute-resolution",
+        "work_type": "analyze",
+        "n_criteria": 46,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.8043,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/extract-key-terms-from-vendor-proposal",
+        "title": "Extract key terms from vendor proposal",
+        "area": "intellectual-property",
+        "work_type": "analyze",
+        "n_criteria": 67,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.806,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/analyze-change-of-control-provisions-across-targets-material-contracts",
+        "title": "Analyze change of control provisions across targets material contracts",
+        "area": "corporate-ma",
+        "work_type": "analyze",
+        "n_criteria": 57,
+        "n_docs": 19,
+        "n_deliverables": 1,
+        "pooled": 0.807,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/analyze-restrictive-covenant-enforceability-across-six-employment-agreements",
+        "title": "Analyze restrictive covenant enforceability across six employment agreements",
+        "area": "intellectual-property",
+        "work_type": "analyze",
+        "n_criteria": 59,
+        "n_docs": 9,
+        "n_deliverables": 1,
+        "pooled": 0.8136,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "healthcare-life-sciences/draft-compliance-training-manual",
+        "title": "Draft compliance training manual",
+        "area": "healthcare-life-sciences",
+        "work_type": "draft",
+        "n_criteria": 72,
+        "n_docs": 8,
+        "n_deliverables": 1,
+        "pooled": 0.8194,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "litigation-dispute-resolution/draft-motion-to-dismiss-brief",
+        "title": "Draft motion to dismiss brief",
+        "area": "litigation-dispute-resolution",
+        "work_type": "draft",
+        "n_criteria": 89,
+        "n_docs": 24,
+        "n_deliverables": 3,
+        "pooled": 0.8202,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "data-privacy-cybersecurity/review-incident-response-plan-against-regulatory-requirements-and-industry-standards",
+        "title": "Review incident response plan against regulatory requirements and industry standards",
+        "area": "data-privacy-cybersecurity",
+        "work_type": "review",
+        "n_criteria": 39,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.8205,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "funds-asset-management/analyze-fund-economics-comparison",
+        "title": "Analyze fund economics comparison",
+        "area": "funds-asset-management",
+        "work_type": "analyze",
+        "n_criteria": 106,
+        "n_docs": 14,
+        "n_deliverables": 5,
+        "pooled": 0.8208,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "trusts-estates-private-client/draft-prenuptial-agreement",
+        "title": "Draft prenuptial agreement",
+        "area": "trusts-estates-private-client",
+        "work_type": "draft",
+        "n_criteria": 73,
+        "n_docs": 8,
+        "n_deliverables": 2,
+        "pooled": 0.8356,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "environmental-esg/compare-esg-disclosure-against-regulatory-requirements",
+        "title": "Compare esg disclosure against regulatory requirements",
+        "area": "environmental-esg",
+        "work_type": "analyze",
+        "n_criteria": 51,
+        "n_docs": 9,
+        "n_deliverables": 1,
+        "pooled": 0.8431,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "real-estate/identify-issues-in-counterpartys-proposed-commercial-real-estate-loan-agreement",
+        "title": "Identify issues in counterpartys proposed commercial real estate loan agreement",
+        "area": "real-estate",
+        "work_type": "review",
+        "n_criteria": 39,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.8462,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "data-privacy-cybersecurity/review-counterparty-data-processing-agreement",
+        "title": "Review counterparty data processing agreement",
+        "area": "data-privacy-cybersecurity",
+        "work_type": "review",
+        "n_criteria": 46,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.8478,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/draft-motion-to-compel-production-of-source-code-and-technical-documents",
+        "title": "Draft motion to compel production of source code and technical documents",
+        "area": "intellectual-property",
+        "work_type": "draft",
+        "n_criteria": 62,
+        "n_docs": 9,
+        "n_deliverables": 1,
+        "pooled": 0.8548,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "trusts-estates-private-client/draft-estate-accounting-report",
+        "title": "Draft estate accounting report",
+        "area": "trusts-estates-private-client",
+        "work_type": "draft",
+        "n_criteria": 86,
+        "n_docs": 11,
+        "n_deliverables": 1,
+        "pooled": 0.8605,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "environmental-esg/draft-markup-of-environmental-indemnity-agreement",
+        "title": "Draft markup of environmental indemnity agreement",
+        "area": "environmental-esg",
+        "work_type": "draft",
+        "n_criteria": 72,
+        "n_docs": 7,
+        "n_deliverables": 2,
+        "pooled": 0.8611,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "employment-labor/research-non",
+        "title": "Research non",
+        "area": "employment-labor",
+        "work_type": "research",
+        "n_criteria": 45,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.8667,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/synthesize-disclosure-schedules-against-diligence-findings",
+        "title": "Synthesize disclosure schedules against diligence findings",
+        "area": "corporate-ma",
+        "work_type": "analyze",
+        "n_criteria": 47,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.8723,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "antitrust-competition/draft-consent-decree",
+        "title": "Draft consent decree",
+        "area": "antitrust-competition",
+        "work_type": "draft",
+        "n_criteria": 72,
+        "n_docs": 7,
+        "n_deliverables": 2,
+        "pooled": 0.875,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "antitrust-competition/extract-relevant-product-market-definitions-from-precedent-decisions",
+        "title": "Extract relevant product market definitions from precedent decisions",
+        "area": "antitrust-competition",
+        "work_type": "research",
+        "n_criteria": 50,
+        "n_docs": 8,
+        "n_deliverables": 2,
+        "pooled": 0.88,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "healthcare-life-sciences/extract-regulatory-compliance-obligations",
+        "title": "Extract regulatory compliance obligations",
+        "area": "healthcare-life-sciences",
+        "work_type": "analyze",
+        "n_criteria": 50,
+        "n_docs": 7,
+        "n_deliverables": 2,
+        "pooled": 0.88,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "real-estate/draft-loan-agreement",
+        "title": "Draft loan agreement",
+        "area": "real-estate",
+        "work_type": "draft",
+        "n_criteria": 84,
+        "n_docs": 7,
+        "n_deliverables": 2,
+        "pooled": 0.881,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "emerging-companies-venture-capital/extract-key-terms-from-merger-agreement",
+        "title": "Extract key terms from merger agreement",
+        "area": "emerging-companies-venture-capital",
+        "work_type": "analyze",
+        "n_criteria": 77,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.8831,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "banking-finance/draft-fee-letter",
+        "title": "Draft fee letter",
+        "area": "banking-finance",
+        "work_type": "draft",
+        "n_criteria": 70,
+        "n_docs": 6,
+        "n_deliverables": 2,
+        "pooled": 0.8857,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "real-estate/analyze-counterparty-markup-of-commercial-lease-agreement",
+        "title": "Analyze counterparty markup of commercial lease agreement",
+        "area": "real-estate",
+        "work_type": "review",
+        "n_criteria": 65,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.8923,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "insurance/compare-commercial-property-policy-form-against-state-regulatory-requirements",
+        "title": "Compare commercial property policy form against state regulatory requirements",
+        "area": "insurance",
+        "work_type": "review",
+        "n_criteria": 38,
+        "n_docs": 4,
+        "n_deliverables": 1,
+        "pooled": 0.8947,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "tax/extract-document-requests-from-tax-authority-summons",
+        "title": "Extract document requests from tax authority summons",
+        "area": "tax",
+        "work_type": "analyze",
+        "n_criteria": 48,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.8958,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "data-privacy-cybersecurity/draft-updated-privacy-policy",
+        "title": "Draft updated privacy policy",
+        "area": "data-privacy-cybersecurity",
+        "work_type": "draft",
+        "n_criteria": 68,
+        "n_docs": 6,
+        "n_deliverables": 2,
+        "pooled": 0.8971,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "trusts-estates-private-client/draft-qualified-domestic-relations-order",
+        "title": "Draft qualified domestic relations order",
+        "area": "trusts-estates-private-client",
+        "work_type": "draft",
+        "n_criteria": 80,
+        "n_docs": 9,
+        "n_deliverables": 3,
+        "pooled": 0.9,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "arbitration-international-dispute-resolution/draft-expert-report-on-construction-delay-and-damages",
+        "title": "Draft expert report on construction delay and damages",
+        "area": "arbitration-international-dispute-resolution",
+        "work_type": "draft",
+        "n_criteria": 71,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.9014,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "capital-markets/review-form-8",
+        "title": "Review form 8",
+        "area": "capital-markets",
+        "work_type": "review",
+        "n_criteria": 32,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.9062,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/compare-post",
+        "title": "Compare post",
+        "area": "corporate-ma",
+        "work_type": "analyze",
+        "n_criteria": 45,
+        "n_docs": 6,
+        "n_deliverables": 1,
+        "pooled": 0.9111,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/draft-markup-of-stock-purchase-agreement-provisions",
+        "title": "Draft markup of stock purchase agreement provisions",
+        "area": "corporate-ma",
+        "work_type": "review",
+        "n_criteria": 59,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.9153,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "white-collar-defense-investigations/draft-response-to-sec-wells-notice",
+        "title": "Draft response to sec wells notice",
+        "area": "white-collar-defense-investigations",
+        "work_type": "draft",
+        "n_criteria": 40,
+        "n_docs": 10,
+        "n_deliverables": 1,
+        "pooled": 0.925,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "healthcare-life-sciences/draft-enterprise-saas-vendor-onboarding-due-diligence",
+        "title": "Draft enterprise saas vendor onboarding due diligence",
+        "area": "healthcare-life-sciences",
+        "work_type": "draft",
+        "n_criteria": 56,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.9286,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/extract-key-terms-from-patent-specification",
+        "title": "Extract key terms from patent specification",
+        "area": "intellectual-property",
+        "work_type": "analyze",
+        "n_criteria": 59,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.9322,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "intellectual-property/draft-source-code-escrow-agreement",
+        "title": "Draft source code escrow agreement",
+        "area": "intellectual-property",
+        "work_type": "draft",
+        "n_criteria": 60,
+        "n_docs": 6,
+        "n_deliverables": 2,
+        "pooled": 0.9333,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-ma/draft-issues-list-for-transition-services-agreement-tsa",
+        "title": "Draft issues list for transition services agreement tsa",
+        "area": "corporate-ma",
+        "work_type": "review",
+        "n_criteria": 64,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.9375,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "trusts-estates-private-client/compare-parenting-plan-against-state-guidelines",
+        "title": "Compare parenting plan against state guidelines",
+        "area": "trusts-estates-private-client",
+        "work_type": "analyze",
+        "n_criteria": 51,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.9412,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "international-trade-sanctions/draft-cfius-voluntary-notice",
+        "title": "Draft cfius voluntary notice",
+        "area": "international-trade-sanctions",
+        "work_type": "draft",
+        "n_criteria": 71,
+        "n_docs": 7,
+        "n_deliverables": 1,
+        "pooled": 0.9437,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-governance/draft-corporate-code-of-conduct",
+        "title": "Draft corporate code of conduct",
+        "area": "corporate-governance",
+        "work_type": "draft",
+        "n_criteria": 76,
+        "n_docs": 10,
+        "n_deliverables": 2,
+        "pooled": 0.9474,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "energy-natural-resources/compare-site-lease-against-market-terms",
+        "title": "Compare site lease against market terms",
+        "area": "energy-natural-resources",
+        "work_type": "analyze",
+        "n_criteria": 53,
+        "n_docs": 5,
+        "n_deliverables": 1,
+        "pooled": 0.9811,
+        "all_pass_rate": 0.0
+      },
+      {
+        "task": "corporate-governance/map-beneficial-ownership-reporting-requirements-to-corporate-entity-structure",
+        "title": "Map beneficial ownership reporting requirements to corporate entity structure",
+        "area": "corporate-governance",
+        "work_type": "analyze",
+        "n_criteria": 60,
+        "n_docs": 8,
+        "n_deliverables": 1,
+        "pooled": 0.9833,
+        "all_pass_rate": 0.0
+      }
+    ]
+  }
+}

data/lab_candidate_landscape.json

@@ -0,0 +1,237 @@
+{
+  "model": "DeepSeek-V4-Pro-together",
+  "subset": "dev",
+  "trials": 3,
+  "candidates": [
+    {
+      "harness": "matter_audit_allwork",
+      "mechanism": "matter audit (best-of-N vote) on draft+analyze+review",
+      "type": "code",
+      "status": "frontier",
+      "pooled": 0.833,
+      "all_pass": 0.0423,
+      "mean_tokens": 2173300,
+      "blended": 0.8432
+    },
+    {
+      "harness": "multi_deliverable_landing_gate",
+      "mechanism": "landing gate hoists a misplaced deliverable (multi-deliverable)",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.8226,
+      "all_pass": 0.0556,
+      "mean_tokens": 2029615,
+      "blended": 0.8402
+    },
+    {
+      "harness": "deliverable_superset_gate",
+      "mechanism": "land the fuller superset of a multi-file deliverable",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.8339,
+      "all_pass": 0.0278,
+      "mean_tokens": 2089408,
+      "blended": 0.8373
+    },
+    {
+      "harness": "deliverable_reassembly_gate",
+      "mechanism": "reassemble a chunked-write-clobbered deliverable",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.8355,
+      "all_pass": 0.0139,
+      "mean_tokens": 2038154,
+      "blended": 0.8322
+    },
+    {
+      "harness": "matter_audit_quorum",
+      "mechanism": "matter audit via JSON-only majority vote (draft-only)",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.823,
+      "all_pass": 0.0429,
+      "mean_tokens": 2572882,
+      "blended": 0.8315
+    },
+    {
+      "harness": "markup_loop_guard",
+      "mechanism": "markup/redline playbook + repetition loop-guard",
+      "type": "mixed",
+      "status": "frontier",
+      "pooled": 0.8273,
+      "all_pass": 0.0278,
+      "mean_tokens": 2149011,
+      "blended": 0.8305
+    },
+    {
+      "harness": "markup_review_playbook",
+      "mechanism": "redline/markup deliverable-structure prompt",
+      "type": "prompt",
+      "status": "explored",
+      "pooled": 0.7981,
+      "all_pass": 0.0278,
+      "mean_tokens": 2247545,
+      "blended": 0.8008
+    },
+    {
+      "harness": "matter_fidelity_finish_guard",
+      "mechanism": "matter-fidelity self-check + finish guard",
+      "type": "code",
+      "status": "frontier",
+      "pooled": 0.8058,
+      "all_pass": 0.0,
+      "mean_tokens": 1935758,
+      "blended": 0.7962
+    },
+    {
+      "harness": "deliverable_landing_gate",
+      "mechanism": "land deliverables at the top level of output/",
+      "type": "code",
+      "status": "frontier",
+      "pooled": 0.792,
+      "all_pass": 0.0139,
+      "mean_tokens": 1875117,
+      "blended": 0.7896
+    },
+    {
+      "harness": "deliverable_finish_guard",
+      "mechanism": "finish-hook against an empty output/",
+      "type": "code",
+      "status": "frontier",
+      "pooled": 0.7871,
+      "all_pass": 0.0,
+      "mean_tokens": 1983441,
+      "blended": 0.7771
+    },
+    {
+      "harness": "matter_fidelity_loop_breaker",
+      "mechanism": "matter-fidelity + investigation doom-loop breaker",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.7739,
+      "all_pass": 0.0139,
+      "mean_tokens": 2047798,
+      "blended": 0.7706
+    },
+    {
+      "harness": "matter_fidelity_review",
+      "mechanism": "draft-only runtime matter-fidelity self-check",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.7671,
+      "all_pass": 0.0,
+      "mean_tokens": 1911727,
+      "blended": 0.7576
+    },
+    {
+      "harness": "landing_gate_loop_breaker",
+      "mechanism": "landing gate + repetition circuit-breaker",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.7549,
+      "all_pass": 0.0141,
+      "mean_tokens": 1689266,
+      "blended": 0.7534
+    },
+    {
+      "harness": "loop_circuit_breaker",
+      "mechanism": "detect a repeated identical tool call, inject a corrective",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.7251,
+      "all_pass": 0.0139,
+      "mean_tokens": 1689360,
+      "blended": 0.7235
+    },
+    {
+      "harness": "drafting_matter_repair",
+      "mechanism": "lean draft playbook (matter-grounding + record-fidelity)",
+      "type": "prompt",
+      "status": "frontier",
+      "pooled": 0.7278,
+      "all_pass": 0.0139,
+      "mean_tokens": 2299051,
+      "blended": 0.7232
+    },
+    {
+      "harness": "toolcall_json_repair",
+      "mechanism": "repair provider-corrupted tool-call JSON in-loop",
+      "type": "code",
+      "status": "manual",
+      "pooled": 0.7079,
+      "all_pass": 0.0278,
+      "mean_tokens": 1903059,
+      "blended": 0.7122
+    },
+    {
+      "harness": "delivery_assurance_gate",
+      "mechanism": "zero-token post-solve deliverable-assurance gate",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.7047,
+      "all_pass": 0.0139,
+      "mean_tokens": 1707343,
+      "blended": 0.7031
+    },
+    {
+      "harness": "analysis_coverage_playbook",
+      "mechanism": "coverage/precision/issue-lens prompt for analyze+review",
+      "type": "prompt",
+      "status": "frontier",
+      "pooled": 0.6895,
+      "all_pass": 0.0,
+      "mean_tokens": 1755050,
+      "blended": 0.6807
+    },
+    {
+      "harness": "drafting_matter_playbook",
+      "mechanism": "lean drafting prompt for the draft work type",
+      "type": "prompt",
+      "status": "explored",
+      "pooled": 0.6775,
+      "all_pass": 0.0139,
+      "mean_tokens": 1832516,
+      "blended": 0.6753
+    },
+    {
+      "harness": "self_review_revise",
+      "mechanism": "runtime self-review / verify-and-revise of the first draft",
+      "type": "code",
+      "status": "manual",
+      "pooled": 0.6606,
+      "all_pass": 0.0,
+      "mean_tokens": 2462037,
+      "blended": 0.6482
+    },
+    {
+      "harness": "baseline_vanilla",
+      "mechanism": "stock LAB harness",
+      "type": "baseline",
+      "status": "frontier",
+      "pooled": 0.6309,
+      "all_pass": 0.0,
+      "mean_tokens": 1417242,
+      "blended": 0.6238
+    },
+    {
+      "harness": "output_completion_gate",
+      "mechanism": "stop-hook nudging a final write + copy a misplaced deliverable",
+      "type": "code",
+      "status": "explored",
+      "pooled": 0.6194,
+      "all_pass": 0.0,
+      "mean_tokens": 1379475,
+      "blended": 0.6125
+    },
+    {
+      "harness": "issue_spotting_playbook",
+      "mechanism": "analytical-rigor / issue-spotting prompt (heavy, later distilled)",
+      "type": "prompt",
+      "status": "manual",
+      "pooled": 0.5716,
+      "all_pass": 0.0,
+      "mean_tokens": 3518512,
+      "blended": 0.554
+    }
+  ]
+}

data/lab_cross_model_transfer.json

@@ -0,0 +1,56 @@
+{
+  "harness": "deliverable_landing_gate",
+  "transfer": [
+    {
+      "model": "DeepSeek V4 Pro",
+      "model_short": "DeepSeek-V4-Pro-together",
+      "baseline_pooled": 0.6339,
+      "baseline_all_pass": 0.0,
+      "baseline_n": 100,
+      "harness_pooled": 0.7686,
+      "harness_all_pass": 0.02,
+      "harness_n": 100,
+      "delta_pp": 13.5,
+      "evaluable": true
+    },
+    {
+      "model": "DeepSeek V4 Flash",
+      "model_short": "DeepSeek-V4-Flash",
+      "baseline_pooled": 0.6463,
+      "baseline_all_pass": 0.0,
+      "baseline_n": 100,
+      "harness_pooled": 0.79,
+      "harness_all_pass": 0.02,
+      "harness_n": 100,
+      "delta_pp": 14.4,
+      "evaluable": true
+    },
+    {
+      "model": "Nemotron-3 Ultra 550B",
+      "model_short": "NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4",
+      "baseline_pooled": 0.6384,
+      "baseline_all_pass": 0.0,
+      "baseline_n": 100,
+      "harness_pooled": 0.6421,
+      "harness_all_pass": 0.03,
+      "harness_n": 100,
+      "delta_pp": 0.4,
+      "evaluable": true
+    },
+    {
+      "model": "GLM-5.2",
+      "model_short": "GLM-5-2",
+      "baseline_pooled": 0.3446,
+      "baseline_all_pass": 0.02,
+      "baseline_n": 100,
+      "evaluable": false
+    }
+  ],
+  "frontier_on_tuned_model": {
+    "model": "DeepSeek V4 Pro",
+    "harness": "matter_audit_allwork",
+    "pooled": 0.8008,
+    "all_pass": 0.05,
+    "n": 100
+  }
+}

data/lab_external_harnesses.json

@@ -0,0 +1,75 @@
+{
+  "n_test": 100,
+  "rows": [
+    {
+      "harness": "Optimized LAB",
+      "short": "Optimized LAB",
+      "pooled": 0.801,
+      "macro": 0.802,
+      "all_pass": 0.05,
+      "mean_tokens_m": 2.19,
+      "timeout_runs": null,
+      "expected_file_type": null,
+      "note": "final meta-harness frontier (matter_audit_allwork)",
+      "kind": "optimized",
+      "delta_vs_vanilla_pp": 16.7,
+      "delta_vs_frontier_pp": 0.0
+    },
+    {
+      "harness": "Vanilla LAB",
+      "short": "Vanilla LAB",
+      "pooled": 0.634,
+      "macro": 0.65,
+      "all_pass": 0.0,
+      "mean_tokens_m": 1.76,
+      "timeout_runs": null,
+      "expected_file_type": null,
+      "note": "stock LAB loop/tools/skills",
+      "kind": "lab",
+      "delta_vs_vanilla_pp": 0.0,
+      "delta_vs_frontier_pp": -16.7
+    },
+    {
+      "harness": "Pi",
+      "short": "Pi",
+      "pooled": 0.454,
+      "macro": 0.455,
+      "all_pass": 0.0,
+      "mean_tokens_m": null,
+      "timeout_runs": 25,
+      "expected_file_type": 75,
+      "note": "strongest general harness, but 18.0pp below vanilla LAB",
+      "kind": "external",
+      "delta_vs_vanilla_pp": -18.0,
+      "delta_vs_frontier_pp": -34.7
+    },
+    {
+      "harness": "Goose",
+      "short": "Goose",
+      "pooled": 0.232,
+      "macro": 0.226,
+      "all_pass": 0.0,
+      "mean_tokens_m": null,
+      "timeout_runs": 73,
+      "expected_file_type": 26,
+      "note": "frequent timeout and delivery failures",
+      "kind": "external",
+      "delta_vs_vanilla_pp": -40.2,
+      "delta_vs_frontier_pp": -56.9
+    },
+    {
+      "harness": "mini-swe-agent",
+      "short": "mini-swe",
+      "pooled": 0.035,
+      "macro": 0.031,
+      "all_pass": 0.0,
+      "mean_tokens_m": 2.47,
+      "timeout_runs": 13,
+      "expected_file_type": 7,
+      "note": "bash-only SWE scaffold is a poor LAB fit",
+      "kind": "external",
+      "delta_vs_vanilla_pp": -59.9,
+      "delta_vs_frontier_pp": -76.6
+    }
+  ]
+}

data/lab_frontier_climb.json

@@ -0,0 +1,259 @@
+{
+  "model": "DeepSeek-V4-Pro-together",
+  "subset": "dev",
+  "trials": 3,
+  "points": [
+    {
+      "iteration": 0,
+      "harness": "baseline_vanilla",
+      "mechanism": "stock LAB harness",
+      "type": "baseline",
+      "pooled": 0.6309,
+      "all_pass": 0.0,
+      "mean_tokens": 1417242,
+      "promoted": true,
+      "frontier_after": "baseline_vanilla",
+      "frontier_pooled_after": 0.6309
+    },
+    {
+      "iteration": 1,
+      "harness": "output_completion_gate",
+      "mechanism": "stop-hook nudging a final write + copy a misplaced deliverable",
+      "type": "code",
+      "pooled": 0.6194,
+      "all_pass": 0.0,
+      "mean_tokens": 1379475,
+      "promoted": false,
+      "frontier_after": "baseline_vanilla",
+      "frontier_pooled_after": 0.6309
+    },
+    {
+      "iteration": 4,
+      "harness": "analysis_coverage_playbook",
+      "mechanism": "coverage/precision/issue-lens prompt for analyze+review",
+      "type": "prompt",
+      "pooled": 0.6895,
+      "all_pass": 0.0,
+      "mean_tokens": 1755050,
+      "promoted": true,
+      "frontier_after": "analysis_coverage_playbook",
+      "frontier_pooled_after": 0.6895
+    },
+    {
+      "iteration": 5,
+      "harness": "drafting_matter_playbook",
+      "mechanism": "lean drafting prompt for the draft work type",
+      "type": "prompt",
+      "pooled": 0.6775,
+      "all_pass": 0.0139,
+      "mean_tokens": 1832516,
+      "promoted": false,
+      "frontier_after": "analysis_coverage_playbook",
+      "frontier_pooled_after": 0.6895
+    },
+    {
+      "iteration": 6,
+      "harness": "delivery_assurance_gate",
+      "mechanism": "zero-token post-solve deliverable-assurance gate",
+      "type": "code",
+      "pooled": 0.7047,
+      "all_pass": 0.0139,
+      "mean_tokens": 1707343,
+      "promoted": false,
+      "frontier_after": "analysis_coverage_playbook",
+      "frontier_pooled_after": 0.6895
+    },
+    {
+      "iteration": 7,
+      "harness": "toolcall_json_repair",
+      "mechanism": "repair provider-corrupted tool-call JSON in-loop",
+      "type": "code",
+      "pooled": 0.7079,
+      "all_pass": 0.0278,
+      "mean_tokens": 1903059,
+      "promoted": false,
+      "frontier_after": "analysis_coverage_playbook",
+      "frontier_pooled_after": 0.6895
+    },
+    {
+      "iteration": 8,
+      "harness": "drafting_matter_repair",
+      "mechanism": "lean draft playbook (matter-grounding + record-fidelity)",
+      "type": "prompt",
+      "pooled": 0.7278,
+      "all_pass": 0.0139,
+      "mean_tokens": 2299051,
+      "promoted": true,
+      "frontier_after": "drafting_matter_repair",
+      "frontier_pooled_after": 0.7278
+    },
+    {
+      "iteration": 9,
+      "harness": "loop_circuit_breaker",
+      "mechanism": "detect a repeated identical tool call, inject a corrective",
+      "type": "code",
+      "pooled": 0.7251,
+      "all_pass": 0.0139,
+      "mean_tokens": 1689360,
+      "promoted": false,
+      "frontier_after": "drafting_matter_repair",
+      "frontier_pooled_after": 0.7278
+    },
+    {
+      "iteration": 10,
+      "harness": "deliverable_landing_gate",
+      "mechanism": "land deliverables at the top level of output/",
+      "type": "code",
+      "pooled": 0.792,
+      "all_pass": 0.0139,
+      "mean_tokens": 1875117,
+      "promoted": true,
+      "frontier_after": "deliverable_landing_gate",
+      "frontier_pooled_after": 0.792
+    },
+    {
+      "iteration": 11,
+      "harness": "landing_gate_loop_breaker",
+      "mechanism": "landing gate + repetition circuit-breaker",
+      "type": "code",
+      "pooled": 0.7549,
+      "all_pass": 0.0139,
+      "mean_tokens": 1689266,
+      "promoted": false,
+      "frontier_after": "deliverable_landing_gate",
+      "frontier_pooled_after": 0.792
+    },
+    {
+      "iteration": 12,
+      "harness": "matter_fidelity_review",
+      "mechanism": "draft-only runtime matter-fidelity self-check",
+      "type": "code",
+      "pooled": 0.7671,
+      "all_pass": 0.0,
+      "mean_tokens": 1911727,
+      "promoted": false,
+      "frontier_after": "deliverable_landing_gate",
+      "frontier_pooled_after": 0.792
+    },
+    {
+      "iteration": 13,
+      "harness": "deliverable_finish_guard",
+      "mechanism": "finish-hook against an empty output/",
+      "type": "code",
+      "pooled": 0.7871,
+      "all_pass": 0.0,
+      "mean_tokens": 1983441,
+      "promoted": true,
+      "frontier_after": "deliverable_finish_guard",
+      "frontier_pooled_after": 0.792
+    },
+    {
+      "iteration": 14,
+      "harness": "matter_fidelity_finish_guard",
+      "mechanism": "matter-fidelity self-check + finish guard",
+      "type": "code",
+      "pooled": 0.8058,
+      "all_pass": 0.0,
+      "mean_tokens": 1935758,
+      "promoted": true,
+      "frontier_after": "matter_fidelity_finish_guard",
+      "frontier_pooled_after": 0.8058
+    },
+    {
+      "iteration": 15,
+      "harness": "matter_fidelity_loop_breaker",
+      "mechanism": "matter-fidelity + investigation doom-loop breaker",
+      "type": "code",
+      "pooled": 0.7739,
+      "all_pass": 0.0139,
+      "mean_tokens": 2047798,
+      "promoted": false,
+      "frontier_after": "matter_fidelity_finish_guard",
+      "frontier_pooled_after": 0.8058
+    },
+    {
+      "iteration": 16,
+      "harness": "markup_review_playbook",
+      "mechanism": "redline/markup deliverable-structure prompt",
+      "type": "prompt",
+      "pooled": 0.7981,
+      "all_pass": 0.0278,
+      "mean_tokens": 2247545,
+      "promoted": false,
+      "frontier_after": "matter_fidelity_finish_guard",
+      "frontier_pooled_after": 0.8058
+    },
+    {
+      "iteration": 17,
+      "harness": "markup_loop_guard",
+      "mechanism": "markup/redline playbook + repetition loop-guard",
+      "type": "mixed",
+      "pooled": 0.8273,
+      "all_pass": 0.0278,
+      "mean_tokens": 2149011,
+      "promoted": true,
+      "frontier_after": "markup_loop_guard",
+      "frontier_pooled_after": 0.8273
+    },
+    {
+      "iteration": 18,
+      "harness": "matter_audit_quorum",
+      "mechanism": "matter audit via JSON-only majority vote (draft-only)",
+      "type": "code",
+      "pooled": 0.823,
+      "all_pass": 0.0417,
+      "mean_tokens": 2572882,
+      "promoted": false,
+      "frontier_after": "markup_loop_guard",
+      "frontier_pooled_after": 0.8273
+    },
+    {
+      "iteration": 19,
+      "harness": "matter_audit_allwork",
+      "mechanism": "matter audit (best-of-N vote) on draft+analyze+review",
+      "type": "code",
+      "pooled": 0.833,
+      "all_pass": 0.0417,
+      "mean_tokens": 2173300,
+      "promoted": true,
+      "frontier_after": "markup_loop_guard",
+      "frontier_pooled_after": 0.833
+    },
+    {
+      "iteration": 20,
+      "harness": "multi_deliverable_landing_gate",
+      "mechanism": "landing gate hoists a misplaced deliverable (multi-deliverable)",
+      "type": "code",
+      "pooled": 0.8226,
+      "all_pass": 0.0556,
+      "mean_tokens": 2029615,
+      "promoted": false,
+      "frontier_after": "matter_audit_allwork",
+      "frontier_pooled_after": 0.833
+    },
+    {
+      "iteration": 21,
+      "harness": "deliverable_superset_gate",
+      "mechanism": "land the fuller superset of a multi-file deliverable",
+      "type": "code",
+      "pooled": 0.8339,
+      "all_pass": 0.0278,
+      "mean_tokens": 2089408,
+      "promoted": false,
+      "frontier_after": "matter_audit_allwork",
+      "frontier_pooled_after": 0.833
+    },
+    {
+      "iteration": 22,
+      "harness": "deliverable_reassembly_gate",
+      "mechanism": "reassemble a chunked-write-clobbered deliverable",
+      "type": "code",
+      "pooled": 0.8355,
+      "all_pass": 0.0139,
+      "mean_tokens": 2038154,
+      "promoted": false,
+      "frontier_after": "matter_audit_allwork",
+      "frontier_pooled_after": 0.833
+    }
+  ]
+}

data/lab_rescue_heatmap.json

@@ -0,0 +1,686 @@
+{
+  "steps": [
+    {
+      "label": "baseline",
+      "harness": "baseline_vanilla"
+    },
+    {
+      "label": "+ coverage playbook",
+      "harness": "analysis_coverage_playbook"
+    },
+    {
+      "label": "+ drafting repair",
+      "harness": "drafting_matter_repair"
+    },
+    {
+      "label": "+ landing gate",
+      "harness": "deliverable_landing_gate"
+    },
+    {
+      "label": "+ finish guard",
+      "harness": "deliverable_finish_guard"
+    },
+    {
+      "label": "+ matter fidelity",
+      "harness": "matter_fidelity_finish_guard"
+    },
+    {
+      "label": "+ markup loop guard",
+      "harness": "markup_loop_guard"
+    },
+    {
+      "label": "+ matter audit",
+      "harness": "matter_audit_allwork"
+    }
+  ],
+  "tasks": [
+    {
+      "id": "antitrust-competition/draft-pre-notification-briefing-paper",
+      "area": "antitrust-competition",
+      "work_type": "draft",
+      "n_criteria": 60,
+      "est_tokens": 73742,
+      "rates": [
+        0.8444,
+        0.8333,
+        0.6278,
+        0.9278,
+        0.9167,
+        0.8889,
+        0.8722,
+        0.9444
+      ],
+      "out_tokens": [
+        13218,
+        18888,
+        20518,
+        27639,
+        19148,
+        18659,
+        66524,
+        42506
+      ]
+    },
+    {
+      "id": "arbitration-international-dispute-resolution/analyze-arbitration-award-for-new-york-convention-enforcement-defenses",
+      "area": "arbitration-international-dispute-resolution",
+      "work_type": "analyze",
+      "n_criteria": 43,
+      "est_tokens": 62856,
+      "rates": [
+        0.5349,
+        0.8062,
+        0.845,
+        0.8372,
+        0.814,
+        0.845,
+        0.8992,
+        0.8915
+      ],
+      "out_tokens": [
+        8639,
+        13697,
+        10487,
+        10745,
+        16444,
+        10903,
+        13421,
+        44524
+      ]
+    },
+    {
+      "id": "banking-finance/identify-term-sheet-issues",
+      "area": "banking-finance",
+      "work_type": "review",
+      "n_criteria": 34,
+      "est_tokens": 23812,
+      "rates": [
+        0.4706,
+        0.7059,
+        0.7353,
+        0.5196,
+        0.6569,
+        0.6863,
+        0.6765,
+        0.7157
+      ],
+      "out_tokens": [
+        8489,
+        22647,
+        17559,
+        13647,
+        11618,
+        18899,
+        17259,
+        18902
+      ]
+    },
+    {
+      "id": "bankruptcy-restructuring/extract-loan-agreement-terms/scenario-02",
+      "area": "bankruptcy-restructuring",
+      "work_type": "analyze",
+      "n_criteria": 102,
+      "est_tokens": 110927,
+      "rates": [
+        0.7451,
+        0.8268,
+        0.5719,
+        0.8235,
+        0.8562,
+        0.8105,
+        0.8758,
+        0.8529
+      ],
+      "out_tokens": [
+        null,
+        null,
+        null,
+        null,
+        null,
+        null,
+        null,
+        null
+      ]
+    },
+    {
+      "id": "capital-markets/analyze-counterparty-markup-of-underwriting-agreement",
+      "area": "capital-markets",
+      "work_type": "analyze",
+      "n_criteria": 58,
+      "est_tokens": 73275,
+      "rates": [
+        0.9483,
+        0.9483,
+        0.9655,
+        0.9425,
+        0.9655,
+        0.9655,
+        0.9828,
+        0.9885
+      ],
+      "out_tokens": [
+        13037,
+        27157,
+        12571,
+        15493,
+        12056,
+        25701,
+        36545,
+        48017
+      ]
+    },
+    {
+      "id": "corporate-governance/research-regulatory-approval-requirements-for-new-fintech-lending-business-line",
+      "area": "corporate-governance",
+      "work_type": "research",
+      "n_criteria": 52,
+      "est_tokens": 75096,
+      "rates": [
+        0.5513,
+        0.5769,
+        0.9038,
+        0.8397,
+        0.891,
+        0.8654,
+        0.9231,
+        0.891
+      ],
+      "out_tokens": [
+        9714,
+        8606,
+        12956,
+        13354,
+        12130,
+        16916,
+        26506,
+        42973
+      ]
+    },
+    {
+      "id": "corporate-ma/draft-restrictive-covenant-agreement",
+      "area": "corporate-ma",
+      "work_type": "draft",
+      "n_criteria": 62,
+      "est_tokens": 32899,
+      "rates": [
+        0.5376,
+        0.7527,
+        0.7849,
+        0.8065,
+        0.8065,
+        0.8011,
+        0.8817,
+        0.8495
+      ],
+      "out_tokens": [
+        15775,
+        15956,
+        14615,
+        17026,
+        12160,
+        12895,
+        19285,
+        22362
+      ]
+    },
+    {
+      "id": "data-privacy-cybersecurity/identify-issues-in-transfer-impact-assessment",
+      "area": "data-privacy-cybersecurity",
+      "work_type": "review",
+      "n_criteria": 33,
+      "est_tokens": 62330,
+      "rates": [
+        0.697,
+        0.798,
+        0.7778,
+        0.5152,
+        0.7475,
+        0.7778,
+        0.7778,
+        0.8485
+      ],
+      "out_tokens": [
+        7350,
+        16352,
+        10994,
+        18340,
+        18546,
+        10948,
+        44021,
+        34717
+      ]
+    },
+    {
+      "id": "emerging-companies-venture-capital/identify-issues-in-investors-rights-agreement",
+      "area": "emerging-companies-venture-capital",
+      "work_type": "review",
+      "n_criteria": 37,
+      "est_tokens": 57671,
+      "rates": [
+        0.4595,
+        0.7568,
+        0.7568,
+        0.7387,
+        0.7838,
+        0.7658,
+        0.8559,
+        0.8108
+      ],
+      "out_tokens": [
+        17438,
+        10663,
+        13677,
+        15845,
+        14579,
+        18883,
+        55119,
+        40910
+      ]
+    },
+    {
+      "id": "employment-labor/compare-non",
+      "area": "employment-labor",
+      "work_type": "analyze",
+      "n_criteria": 38,
+      "est_tokens": 27060,
+      "rates": [
+        0.7982,
+        0.886,
+        0.8158,
+        0.7544,
+        0.5439,
+        0.8333,
+        0.8772,
+        0.8684
+      ],
+      "out_tokens": [
+        11068,
+        10252,
+        9312,
+        10827,
+        8575,
+        15236,
+        14768,
+        12565
+      ]
+    },
+    {
+      "id": "energy-natural-resources/draft-power-purchase-agreement",
+      "area": "energy-natural-resources",
+      "work_type": "draft",
+      "n_criteria": 91,
+      "est_tokens": 101863,
+      "rates": [
+        0.4579,
+        0.8315,
+        0.8681,
+        0.9158,
+        0.8535,
+        0.8535,
+        0.7839,
+        0.5897
+      ],
+      "out_tokens": [
+        52599,
+        43619,
+        81190,
+        53586,
+        41172,
+        46018,
+        70153,
+        130791
+      ]
+    },
+    {
+      "id": "environmental-esg/draft-markup-of-administrative-settlement-agreement",
+      "area": "environmental-esg",
+      "work_type": "review",
+      "n_criteria": 53,
+      "est_tokens": 47705,
+      "rates": [
+        0.0,
+        0.4025,
+        0.2516,
+        0.7862,
+        0.3962,
+        0.4654,
+        0.9686,
+        0.9245
+      ],
+      "out_tokens": [
+        53876,
+        47352,
+        29743,
+        41540,
+        40253,
+        25217,
+        40103,
+        34248
+      ]
+    },
+    {
+      "id": "funds-asset-management/draft-compliance-manual",
+      "area": "funds-asset-management",
+      "work_type": "draft",
+      "n_criteria": 101,
+      "est_tokens": 132869,
+      "rates": [
+        0.6106,
+        0.5677,
+        0.5809,
+        0.8878,
+        0.8449,
+        0.8713,
+        0.6601,
+        0.6403
+      ],
+      "out_tokens": [
+        25735,
+        45973,
+        42205,
+        28411,
+        35140,
+        48270,
+        53665,
+        36361
+      ]
+    },
+    {
+      "id": "healthcare-life-sciences/draft-management-services-agreement",
+      "area": "healthcare-life-sciences",
+      "work_type": "draft",
+      "n_criteria": 94,
+      "est_tokens": 55215,
+      "rates": [
+        0.8865,
+        0.5957,
+        0.922,
+        0.9007,
+        0.9043,
+        0.9291,
+        0.9433,
+        0.922
+      ],
+      "out_tokens": [
+        19031,
+        18272,
+        22192,
+        19340,
+        21652,
+        21476,
+        30743,
+        40985
+      ]
+    },
+    {
+      "id": "immigration/identify-compliance-issues-in-employee-i",
+      "area": "immigration",
+      "work_type": "review",
+      "n_criteria": 39,
+      "est_tokens": 33147,
+      "rates": [
+        0.7521,
+        0.7692,
+        0.8034,
+        0.8632,
+        0.8205,
+        0.812,
+        0.8547,
+        0.8462
+      ],
+      "out_tokens": [
+        12659,
+        12115,
+        23525,
+        11159,
+        14706,
+        10613,
+        32049,
+        24006
+      ]
+    },
+    {
+      "id": "insurance/draft-change-of-control-application",
+      "area": "insurance",
+      "work_type": "draft",
+      "n_criteria": 72,
+      "est_tokens": 100872,
+      "rates": [
+        0.4769,
+        0.7454,
+        0.787,
+        0.8565,
+        0.9028,
+        0.8333,
+        0.963,
+        0.9769
+      ],
+      "out_tokens": [
+        11383,
+        21393,
+        30243,
+        18716,
+        19408,
+        19680,
+        41115,
+        36006
+      ]
+    },
+    {
+      "id": "intellectual-property/research-trade-secret-protections-for-departing-employee",
+      "area": "intellectual-property",
+      "work_type": "research",
+      "n_criteria": 62,
+      "est_tokens": 83276,
+      "rates": [
+        0.8118,
+        0.8118,
+        0.7581,
+        0.8011,
+        0.7957,
+        0.8226,
+        0.8978,
+        0.914
+      ],
+      "out_tokens": [
+        12848,
+        12691,
+        10173,
+        13530,
+        8743,
+        16837,
+        20825,
+        59312
+      ]
+    },
+    {
+      "id": "international-trade-sanctions/compare-entity-details-against-export-control-classification-lists",
+      "area": "international-trade-sanctions",
+      "work_type": "analyze",
+      "n_criteria": 47,
+      "est_tokens": 72041,
+      "rates": [
+        0.8156,
+        0.8227,
+        0.8227,
+        0.8582,
+        0.8227,
+        0.8156,
+        0.8652,
+        0.8511
+      ],
+      "out_tokens": [
+        7376,
+        11294,
+        16173,
+        22147,
+        13838,
+        14951,
+        21597,
+        28757
+      ]
+    },
+    {
+      "id": "litigation-dispute-resolution/draft-motion-for-summary-judgment",
+      "area": "litigation-dispute-resolution",
+      "work_type": "draft",
+      "n_criteria": 69,
+      "est_tokens": 132153,
+      "rates": [
+        0.058,
+        0.0483,
+        0.3237,
+        0.1594,
+        0.2899,
+        0.599,
+        0.4686,
+        0.7295
+      ],
+      "out_tokens": [
+        14947,
+        17341,
+        13032,
+        13719,
+        13230,
+        27618,
+        18033,
+        32511
+      ]
+    },
+    {
+      "id": "real-estate/draft-purchase-and-sale-agreement",
+      "area": "real-estate",
+      "work_type": "draft",
+      "n_criteria": 95,
+      "est_tokens": 104771,
+      "rates": [
+        0.9088,
+        0.8351,
+        0.9193,
+        0.9298,
+        0.9579,
+        0.9544,
+        0.9684,
+        0.9719
+      ],
+      "out_tokens": [
+        21417,
+        18994,
+        18251,
+        19194,
+        18519,
+        30959,
+        39173,
+        28331
+      ]
+    },
+    {
+      "id": "structured-finance-securitization/compare-offering-memorandum-against-indenture",
+      "area": "structured-finance-securitization",
+      "work_type": "review",
+      "n_criteria": 37,
+      "est_tokens": 59772,
+      "rates": [
+        0.7838,
+        0.8108,
+        0.8288,
+        0.8108,
+        0.8198,
+        0.8198,
+        0.8378,
+        0.8514
+      ],
+      "out_tokens": [
+        8895,
+        19424,
+        11980,
+        14166,
+        22312,
+        15507,
+        24678,
+        26067
+      ]
+    },
+    {
+      "id": "tax/analyze-section-382-analysis",
+      "area": "tax",
+      "work_type": "analyze",
+      "n_criteria": 107,
+      "est_tokens": 83796,
+      "rates": [
+        0.4642,
+        0.3302,
+        0.514,
+        0.5919,
+        0.6137,
+        0.567,
+        0.5421,
+        0.6324
+      ],
+      "out_tokens": [
+        17500,
+        20055,
+        21184,
+        24965,
+        24295,
+        19825,
+        50883,
+        81442
+      ]
+    },
+    {
+      "id": "trusts-estates-private-client/identify-issues-in-counterpartys-motion-brief",
+      "area": "trusts-estates-private-client",
+      "work_type": "analyze",
+      "n_criteria": 35,
+      "est_tokens": 40397,
+      "rates": [
+        0.7048,
+        0.7333,
+        0.7238,
+        0.7714,
+        0.6857,
+        0.7619,
+        0.7714,
+        0.8095
+      ],
+      "out_tokens": [
+        7492,
+        14189,
+        10371,
+        15732,
+        11183,
+        10277,
+        16708,
+        19099
+      ]
+    },
+    {
+      "id": "white-collar-defense-investigations/draft-markup-of-deferred-prosecution-agreement",
+      "area": "white-collar-defense-investigations",
+      "work_type": "review",
+      "n_criteria": 52,
+      "est_tokens": 64989,
+      "rates": [
+        0.8397,
+        0.891,
+        0.891,
+        0.9167,
+        0.9231,
+        0.8974,
+        0.9167,
+        0.9487
+      ],
+      "out_tokens": [
+        8823,
+        10812,
+        10701,
+        15323,
+        11223,
+        17324,
+        23002,
+        35196
+      ]
+    }
+  ]
+}

data/lab_summary_numbers.json

@@ -0,0 +1,16 @@
+{
+  "n_iterations": 21,
+  "dev": {
+    "baseline_pooled": 0.6309,
+    "frontier_pooled": 0.833,
+    "frontier_all_pass": 0.0423,
+    "delta_pp": 20.2
+  },
+  "test": {
+    "baseline_pooled": 0.6339,
+    "landing_gate_pooled": 0.7686,
+    "frontier_pooled": 0.8008,
+    "frontier_all_pass": 0.05,
+    "delta_pp": 16.7
+  }
+}

data/lab_trial_variance.json

@@ -0,0 +1,154 @@
+{
+  "n_tasks": 10,
+  "n_trials": 3,
+  "per_trial_pooled": [
+    0.7289,
+    0.7619,
+    0.7491
+  ],
+  "mean_pooled": 0.7466,
+  "std_pp": 1.36,
+  "range_pp": 3.3,
+  "tasks": [
+    {
+      "task": "structured-finance-securitization/compare-offering-memorandum-against-indenture",
+      "title": "Compare offering memorandum against indenture",
+      "area": "structured-finance-securitization",
+      "rates": [
+        0.8378,
+        0.7838,
+        0.0
+      ],
+      "min": 0.0,
+      "max": 0.8378,
+      "mean": 0.5405,
+      "range_pp": 83.8
+    },
+    {
+      "task": "trusts-estates-private-client/identify-issues-in-counterpartys-motion-brief",
+      "title": "Identify issues in counterpartys motion brief",
+      "area": "trusts-estates-private-client",
+      "rates": [
+        0.0,
+        0.7429,
+        0.7429
+      ],
+      "min": 0.0,
+      "max": 0.7429,
+      "mean": 0.4952,
+      "range_pp": 74.3
+    },
+    {
+      "task": "antitrust-competition/draft-pre-notification-briefing-paper",
+      "title": "Draft pre notification briefing paper",
+      "area": "antitrust-competition",
+      "rates": [
+        0.9167,
+        0.7,
+        0.8167
+      ],
+      "min": 0.7,
+      "max": 0.9167,
+      "mean": 0.8111,
+      "range_pp": 21.7
+    },
+    {
+      "task": "banking-finance/identify-term-sheet-issues",
+      "title": "Identify term sheet issues",
+      "area": "banking-finance",
+      "rates": [
+        0.7353,
+        0.5588,
+        0.7059
+      ],
+      "min": 0.5588,
+      "max": 0.7353,
+      "mean": 0.6667,
+      "range_pp": 17.6
+    },
+    {
+      "task": "intellectual-property/research-trade-secret-protections-for-departing-employee",
+      "title": "Research trade secret protections for departing employee",
+      "area": "intellectual-property",
+      "rates": [
+        0.8226,
+        0.7903,
+        0.7097
+      ],
+      "min": 0.7097,
+      "max": 0.8226,
+      "mean": 0.7742,
+      "range_pp": 11.3
+    },
+    {
+      "task": "arbitration-international-dispute-resolution/analyze-arbitration-award-for-new-york-convention-enforcement-defenses",
+      "title": "Analyze arbitration award for new york convention enforcement defenses",
+      "area": "arbitration-international-dispute-resolution",
+      "rates": [
+        0.8837,
+        0.814,
+        0.7907
+      ],
+      "min": 0.7907,
+      "max": 0.8837,
+      "mean": 0.8295,
+      "range_pp": 9.3
+    },
+    {
+      "task": "insurance/draft-change-of-control-application",
+      "title": "Draft change of control application",
+      "area": "insurance",
+      "rates": [
+        0.6944,
+        0.6667,
+        0.75
+      ],
+      "min": 0.6667,
+      "max": 0.75,
+      "mean": 0.7037,
+      "range_pp": 8.3
+    },
+    {
+      "task": "corporate-ma/draft-restrictive-covenant-agreement",
+      "title": "Draft restrictive covenant agreement",
+      "area": "corporate-ma",
+      "rates": [
+        0.7581,
+        0.8226,
+        0.7742
+      ],
+      "min": 0.7581,
+      "max": 0.8226,
+      "mean": 0.7849,
+      "range_pp": 6.5
+    },
+    {
+      "task": "healthcare-life-sciences/draft-management-services-agreement",
+      "title": "Draft management services agreement",
+      "area": "healthcare-life-sciences",
+      "rates": [
+        0.8298,
+        0.8617,
+        0.8936
+      ],
+      "min": 0.8298,
+      "max": 0.8936,
+      "mean": 0.8617,
+      "range_pp": 6.4
+    },
+    {
+      "task": "international-trade-sanctions/compare-entity-details-against-export-control-classification-lists",
+      "title": "Compare entity details against export control classification lists",
+      "area": "international-trade-sanctions",
+      "rates": [
+        0.7234,
+        0.766,
+        0.7447
+      ],
+      "min": 0.7234,
+      "max": 0.766,
+      "mean": 0.7447,
+      "range_pp": 4.3
+    }
+  ]
+}

data/mnist-variant-model.json

@@ -0,0 +1 @@
+{"modelTopology": {"training_config": {"metrics": ["accuracy"], "loss": "categorical_crossentropy", "optimizer_config": {"class_name": "Adam", "config": {"beta_1": 0.8999999761581421, "beta_2": 0.9990000128746033, "decay": 0.0, "epsilon": 1e-07, "amsgrad": false, "lr": 0.0010000000474974513}}, "sample_weight_mode": null, "loss_weights": null}, "keras_version": "2.1.4", "model_config": {"class_name": "Sequential", "config": [{"class_name": "Conv2D", "config": {"kernel_initializer": {"class_name": "VarianceScaling", "config": {"distribution": "uniform", "scale": 1.0, "seed": null, "mode": "fan_avg"}}, "name": "conv2d_1", "kernel_constraint": null, "bias_regularizer": null, "bias_constraint": null, "dtype": "float32", "activation": "linear", "trainable": true, "data_format": "channels_last", "padding": "valid", "strides": [1, 1], "dilation_rate": [1, 1], "kernel_regularizer": null, "filters": 32, "bias_initializer": {"class_name": "Zeros", "config": {}}, "batch_input_shape": [null, 28, 28, 1], "use_bias": true, "activity_regularizer": null, "kernel_size": [3, 3]}}, {"class_name": "Activation", "config": {"activation": "relu", "trainable": true, "name": "activation_1"}}, {"class_name": "Conv2D", "config": {"kernel_initializer": {"class_name": "VarianceScaling", "config": {"distribution": "uniform", "scale": 1.0, "seed": null, "mode": "fan_avg"}}, "name": "conv2d_2", "kernel_constraint": null, "bias_regularizer": null, "bias_constraint": null, "activation": "linear", "trainable": true, "data_format": "channels_last", "padding": "valid", "strides": [1, 1], "dilation_rate": [1, 1], "kernel_regularizer": null, "filters": 32, "bias_initializer": {"class_name": "Zeros", "config": {}}, "use_bias": true, "activity_regularizer": null, "kernel_size": [3, 3]}}, {"class_name": "Activation", "config": {"activation": "relu", "trainable": true, "name": "activation_2"}}, {"class_name": "MaxPooling2D", "config": {"name": "max_pooling2d_1", "trainable": true, "data_format": "channels_last", "pool_size": [2, 2], "padding": "valid", "strides": [2, 2]}}, {"class_name": "Dropout", "config": {"rate": 0.25, "noise_shape": null, "trainable": true, "seed": null, "name": "dropout_1"}}, {"class_name": "Flatten", "config": {"trainable": true, "name": "flatten_1"}}, {"class_name": "Dense", "config": {"kernel_initializer": {"class_name": "VarianceScaling", "config": {"distribution": "uniform", "scale": 1.0, "seed": null, "mode": "fan_avg"}}, "name": "dense_1", "kernel_constraint": null, "bias_regularizer": null, "bias_constraint": null, "activation": "linear", "trainable": true, "kernel_regularizer": null, "bias_initializer": {"class_name": "Zeros", "config": {}}, "units": 128, "use_bias": true, "activity_regularizer": null}}, {"class_name": "Activation", "config": {"activation": "relu", "trainable": true, "name": "activation_3"}}, {"class_name": "Dropout", "config": {"rate": 0.5, "noise_shape": null, "trainable": true, "seed": null, "name": "dropout_2"}}, {"class_name": "Dense", "config": {"kernel_initializer": {"class_name": "VarianceScaling", "config": {"distribution": "uniform", "scale": 1.0, "seed": null, "mode": "fan_avg"}}, "name": "dense_2", "kernel_constraint": null, "bias_regularizer": null, "bias_constraint": null, "activation": "linear", "trainable": true, "kernel_regularizer": null, "bias_initializer": {"class_name": "Zeros", "config": {}}, "units": 5, "use_bias": true, "activity_regularizer": null}}, {"class_name": "Activation", "config": {"activation": "softmax", "trainable": true, "name": "activation_4"}}]}, "backend": "tensorflow"}, "weightsManifest": [{"paths": ["group1-shard1of1"], "weights": [{"dtype": "float32", "shape": [3, 3, 1, 32], "name": "conv2d_1/kernel"}, {"dtype": "float32", "shape": [32], "name": "conv2d_1/bias"}]}, {"paths": ["group2-shard1of1"], "weights": [{"dtype": "float32", "shape": [3, 3, 32, 32], "name": "conv2d_2/kernel"}, {"dtype": "float32", "shape": [32], "name": "conv2d_2/bias"}]}, {"paths": ["group3-shard1of1"], "weights": [{"dtype": "float32", "shape": [4608, 128], "name": "dense_1/kernel"}, {"dtype": "float32", "shape": [128], "name": "dense_1/bias"}]}, {"paths": ["group4-shard1of1"], "weights": [{"dtype": "float32", "shape": [128, 5], "name": "dense_2/kernel"}, {"dtype": "float32", "shape": [5], "name": "dense_2/bias"}]}]}

data/relevance_filters.csv

@@ -0,0 +1,1201 @@
+run,step,metric,value,stderr
+Baseline,1000,ai2d_exact_match,0.2548575129533679,0.007843322436924496
+Baseline,1000,average,0.27120689295763617,
+Baseline,1000,average_rank,3.1,
+Baseline,1000,chartqa_relaxed_overall,0.3308,0.009411906161401973
+Baseline,1000,docvqa_val_anls,0.3528553494243383,0.005852289239342309
+Baseline,1000,infovqa_val_anls,0.17320578642581314,0.006297063452679795
+Baseline,1000,mme_total_score,977.4280712284914,
+Baseline,1000,mmmu_val_mmmu_acc,0.25222,
+Baseline,1000,mmstar_average,0.23215874078908072,
+Baseline,1000,ocrbench_ocrbench_accuracy,0.286,
+Baseline,1000,seedbench_seed_all,0.2563646470261256,
+Baseline,1000,textvqa_val_exact_match,0.3024,0.00628900296642181
+Baseline,2000,ai2d_exact_match,0.26295336787564766,0.007923526907377255
+Baseline,2000,average,0.3202068275596269,
+Baseline,2000,average_rank,2.9,
+Baseline,2000,chartqa_relaxed_overall,0.4688,0.009982508912777261
+Baseline,2000,docvqa_val_anls,0.4452261510942785,0.00614755494712251
+Baseline,2000,infovqa_val_anls,0.1820547866557169,0.006217861455795791
+Baseline,2000,mme_total_score,1049.3036214485794,
+Baseline,2000,mmmu_val_mmmu_acc,0.24556,
+Baseline,2000,mmstar_average,0.21305462434540698,
+Baseline,2000,ocrbench_ocrbench_accuracy,0.395,
+Baseline,2000,seedbench_seed_all,0.258532518065592,
+Baseline,2000,textvqa_val_exact_match,0.41068000000000005,0.006697862330024289
+Baseline,3000,ai2d_exact_match,0.25226683937823835,0.007816909588794397
+Baseline,3000,average,0.3507423834414229,
+Baseline,3000,average_rank,2.7,
+Baseline,3000,chartqa_relaxed_overall,0.5028,0.010001843767601082
+Baseline,3000,docvqa_val_anls,0.502653993831009,0.006267072346683124
+Baseline,3000,infovqa_val_anls,0.21728617578189535,0.006796941784959762
+Baseline,3000,mme_total_score,1170.2383953581434,
+Baseline,3000,mmmu_val_mmmu_acc,0.27556,
+Baseline,3000,mmstar_average,0.25432376938577683,
+Baseline,3000,ocrbench_ocrbench_accuracy,0.436,
+Baseline,3000,seedbench_seed_all,0.2792106725958866,
+Baseline,3000,textvqa_val_exact_match,0.43658,0.006766885462882726
+Baseline,4000,ai2d_exact_match,0.2645725388601036,0.007939149662089447
+Baseline,4000,average,0.36961781722974835,
+Baseline,4000,average_rank,3.7,
+Baseline,4000,chartqa_relaxed_overall,0.5312,0.009982508912777261
+Baseline,4000,docvqa_val_anls,0.5374434618615119,0.0062905728113059655
+Baseline,4000,infovqa_val_anls,0.2287924838861707,0.006994568698639919
+Baseline,4000,mme_total_score,1155.203781512605,
+Baseline,4000,mmmu_val_mmmu_acc,0.25556,
+Baseline,4000,mmstar_average,0.2575590188757354,
+Baseline,4000,ocrbench_ocrbench_accuracy,0.453,
+Baseline,4000,seedbench_seed_all,0.33913285158421347,
+Baseline,4000,textvqa_val_exact_match,0.4593,0.006791695475025738
+Baseline,5000,ai2d_exact_match,0.3125,0.008342439145556371
+Baseline,5000,average,0.3974627910380972,
+Baseline,5000,average_rank,3.3,
+Baseline,5000,chartqa_relaxed_overall,0.5488,0.00995424828018316
+Baseline,5000,docvqa_val_anls,0.552360266782429,0.006300308519952055
+Baseline,5000,infovqa_val_anls,0.23425555286643698,0.007002254622066442
+Baseline,5000,mme_total_score,1181.4653861544618,
+Baseline,5000,mmmu_val_mmmu_acc,0.26667,
+Baseline,5000,mmstar_average,0.29596648146165705,
+Baseline,5000,ocrbench_ocrbench_accuracy,0.462,
+Baseline,5000,seedbench_seed_all,0.43107281823235133,
+Baseline,5000,textvqa_val_exact_match,0.47354000000000007,0.0068172185364497985
+Baseline,6000,ai2d_exact_match,0.358160621761658,0.008629463221867162
+Baseline,6000,average,0.4161227404571003,
+Baseline,6000,average_rank,2.6,
+Baseline,6000,chartqa_relaxed_overall,0.5628,0.00992279440175477
+Baseline,6000,docvqa_val_anls,0.5747451497228876,0.00625495440870239
+Baseline,6000,infovqa_val_anls,0.22152017368968838,0.006604546680525351
+Baseline,6000,mme_total_score,1284.1648659463785,
+Baseline,6000,mmmu_val_mmmu_acc,0.27111,
+Baseline,6000,mmstar_average,0.2978489412854164,
+Baseline,6000,ocrbench_ocrbench_accuracy,0.495,
+Baseline,6000,seedbench_seed_all,0.4795997776542524,
+Baseline,6000,textvqa_val_exact_match,0.48432,0.006800535050670284
+Baseline,7000,ai2d_exact_match,0.3707901554404145,0.00869347755587734
+Baseline,7000,average,0.4291083177345374,
+Baseline,7000,average_rank,2.9,
+Baseline,7000,chartqa_relaxed_overall,0.5656,0.009915542506251351
+Baseline,7000,docvqa_val_anls,0.5940907049431567,0.006224236305767187
+Baseline,7000,infovqa_val_anls,0.2515675215816963,0.007105097396092786
+Baseline,7000,mme_total_score,1185.875650260104,
+Baseline,7000,mmmu_val_mmmu_acc,0.26556,
+Baseline,7000,mmstar_average,0.31372400960777047,
+Baseline,7000,ocrbench_ocrbench_accuracy,0.504,
+Baseline,7000,seedbench_seed_all,0.4964424680377988,
+Baseline,7000,textvqa_val_exact_match,0.5002,0.006794794025220267
+Baseline,8000,ai2d_exact_match,0.37759067357512954,0.008725299846043883
+Baseline,8000,average,0.43846759477995995,
+Baseline,8000,average_rank,3.2,
+Baseline,8000,chartqa_relaxed_overall,0.5832,0.009862556058385773
+Baseline,8000,docvqa_val_anls,0.6017336419437208,0.006231612198089698
+Baseline,8000,infovqa_val_anls,0.2449256624147254,0.006992518502948913
+Baseline,8000,mme_total_score,1199.2409963985594,
+Baseline,8000,mmmu_val_mmmu_acc,0.28111,
+Baseline,8000,mmstar_average,0.33512257186205047,
+Baseline,8000,ocrbench_ocrbench_accuracy,0.51,
+Baseline,8000,seedbench_seed_all,0.5024458032240133,
+Baseline,8000,textvqa_val_exact_match,0.51008,0.006796301690135059
+Baseline,9000,ai2d_exact_match,0.4067357512953368,0.008841214921078996
+Baseline,9000,average,0.4422510732201056,
+Baseline,9000,average_rank,3.2,
+Baseline,9000,chartqa_relaxed_overall,0.5912,0.009834211136815875
+Baseline,9000,docvqa_val_anls,0.6170968481662739,0.00617235763542544
+Baseline,9000,infovqa_val_anls,0.23537031288570615,0.00670318154156447
+Baseline,9000,mme_total_score,1231.5195078031213,
+Baseline,9000,mmmu_val_mmmu_acc,0.25889,
+Baseline,9000,mmstar_average,0.3216444898242951,
+Baseline,9000,ocrbench_ocrbench_accuracy,0.515,
+Baseline,9000,seedbench_seed_all,0.5120622568093385,
+Baseline,9000,textvqa_val_exact_match,0.52226,0.006792711289708482
+Baseline,10000,ai2d_exact_match,0.39993523316062174,0.008817096257082848
+Baseline,10000,average,0.4523875703250908,
+Baseline,10000,average_rank,2.9,
+Baseline,10000,chartqa_relaxed_overall,0.5996,0.00980154906867574
+Baseline,10000,docvqa_val_anls,0.6262613496433054,0.006147756371688175
+Baseline,10000,infovqa_val_anls,0.263290074230132,0.007186788766942786
+Baseline,10000,mme_total_score,1240.8218287314926,
+Baseline,10000,mmmu_val_mmmu_acc,0.28778,
+Baseline,10000,mmstar_average,0.32972717906018517,
+Baseline,10000,ocrbench_ocrbench_accuracy,0.517,
+Baseline,10000,seedbench_seed_all,0.5217342968315731,
+Baseline,10000,textvqa_val_exact_match,0.5261600000000001,0.006785774843600811
+Baseline,11000,ai2d_exact_match,0.422279792746114,0.008889771831066474
+Baseline,11000,average,0.4561398159525099,
+Baseline,11000,average_rank,3.0,
+Baseline,11000,chartqa_relaxed_overall,0.6104,0.009755142291143075
+Baseline,11000,docvqa_val_anls,0.6373130149166712,0.006128022584995044
+Baseline,11000,infovqa_val_anls,0.24419378339723755,0.006897644885887063
+Baseline,11000,mme_total_score,1322.9488795518205,
+Baseline,11000,mmmu_val_mmmu_acc,0.27778,
+Baseline,11000,mmstar_average,0.3298563439522548,
+Baseline,11000,ocrbench_ocrbench_accuracy,0.521,
+Baseline,11000,seedbench_seed_all,0.5237354085603113,
+Baseline,11000,textvqa_val_exact_match,0.5387,0.006770851562852138
+Baseline,12000,ai2d_exact_match,0.42001295336787564,0.008883255931688034
+Baseline,12000,average,0.4582751140055433,
+Baseline,12000,average_rank,3.5,
+Baseline,12000,chartqa_relaxed_overall,0.618,0.009719474639861454
+Baseline,12000,docvqa_val_anls,0.6393961983751871,0.0061228747388476674
+Baseline,12000,infovqa_val_anls,0.24798874058574302,0.006855374548993139
+Baseline,12000,mme_total_score,1225.6453581432572,
+Baseline,12000,mmmu_val_mmmu_acc,0.27889,
+Baseline,12000,mmstar_average,0.34010867846816534,
+Baseline,12000,ocrbench_ocrbench_accuracy,0.512,
+Baseline,12000,seedbench_seed_all,0.5350194552529183,
+Baseline,12000,textvqa_val_exact_match,0.5330600000000001,0.006777713092109446
+Baseline,13000,ai2d_exact_match,0.4375,0.008928571428571428
+Baseline,13000,average,0.4692868662590049,
+Baseline,13000,average_rank,2.7,
+Baseline,13000,chartqa_relaxed_overall,0.6148,0.00973479791861169
+Baseline,13000,docvqa_val_anls,0.6511374872549951,0.006086953065248391
+Baseline,13000,infovqa_val_anls,0.24465055100441893,0.006808432538374664
+Baseline,13000,mme_total_score,1281.7122849139657,
+Baseline,13000,mmmu_val_mmmu_acc,0.28222,
+Baseline,13000,mmstar_average,0.3453069542917521,
+Baseline,13000,ocrbench_ocrbench_accuracy,0.549,
+Baseline,13000,seedbench_seed_all,0.5442468037798777,
+Baseline,13000,textvqa_val_exact_match,0.55472,0.0067416788982325
+Baseline,14000,ai2d_exact_match,0.4572538860103627,0.00896620675297095
+Baseline,14000,average,0.47352486841689195,
+Baseline,14000,average_rank,2.5,
+Baseline,14000,chartqa_relaxed_overall,0.6172,0.009723347231923635
+Baseline,14000,docvqa_val_anls,0.6502269393708169,0.006057950730638126
+Baseline,14000,infovqa_val_anls,0.25805460837190913,0.007037735231659539
+Baseline,14000,mme_total_score,1309.1444577831132,
+Baseline,14000,mmmu_val_mmmu_acc,0.28111,
+Baseline,14000,mmstar_average,0.34575818188776586,
+Baseline,14000,ocrbench_ocrbench_accuracy,0.551,
+Baseline,14000,seedbench_seed_all,0.5483602001111729,
+Baseline,14000,textvqa_val_exact_match,0.55276,0.006751206724612103
+Baseline,15000,ai2d_exact_match,0.45045336787564766,0.008954861634252399
+Baseline,15000,average,0.47878665012878824,
+Baseline,15000,average_rank,2.6,
+Baseline,15000,chartqa_relaxed_overall,0.612,0.009747841205275417
+Baseline,15000,docvqa_val_anls,0.6621413031955148,0.006056838050222495
+Baseline,15000,infovqa_val_anls,0.2706898598157733,0.007200315730154543
+Baseline,15000,mme_total_score,1384.2171868747498,
+Baseline,15000,mmmu_val_mmmu_acc,0.30222,
+Baseline,15000,mmstar_average,0.35408135695920684,
+Baseline,15000,ocrbench_ocrbench_accuracy,0.558,
+Baseline,15000,seedbench_seed_all,0.5411339633129516,
+Baseline,15000,textvqa_val_exact_match,0.5583600000000001,0.0067279027203879065
+Baseline,16000,ai2d_exact_match,0.45077720207253885,0.008955440137395838
+Baseline,16000,average,0.47665128022935843,
+Baseline,16000,average_rank,3.0,
+Baseline,16000,chartqa_relaxed_overall,0.632,0.00964715642305132
+Baseline,16000,docvqa_val_anls,0.6709415729142987,0.005999818105621502
+Baseline,16000,infovqa_val_anls,0.26050032542402035,0.006997451875879188
+Baseline,16000,mme_total_score,1317.8491396558625,
+Baseline,16000,mmmu_val_mmmu_acc,0.27556,
+Baseline,16000,mmstar_average,0.33214333327093315,
+Baseline,16000,ocrbench_ocrbench_accuracy,0.56,
+Baseline,16000,seedbench_seed_all,0.5463590883824346,
+Baseline,16000,textvqa_val_exact_match,0.56158,0.006723854754867398
+Baseline,17000,ai2d_exact_match,0.45919689119170987,0.008969138793675545
+Baseline,17000,average,0.4777141780162423,
+Baseline,17000,average_rank,2.5,
+Baseline,17000,chartqa_relaxed_overall,0.632,0.00964715642305132
+Baseline,17000,docvqa_val_anls,0.6796338519136422,0.005948761388267941
+Baseline,17000,infovqa_val_anls,0.28070956072505215,0.007298333094144192
+Baseline,17000,mme_total_score,1381.9161664665867,
+Baseline,17000,mmmu_val_mmmu_acc,0.27667,
+Baseline,17000,mmstar_average,0.3370289492329521,
+Baseline,17000,ocrbench_ocrbench_accuracy,0.519,
+Baseline,17000,seedbench_seed_all,0.5510283490828238,
+Baseline,17000,textvqa_val_exact_match,0.56416,0.006724830373229479
+Baseline,18000,ai2d_exact_match,0.46567357512953367,0.008977921602780726
+Baseline,18000,average,0.4819834595278701,
+Baseline,18000,average_rank,2.9,
+Baseline,18000,chartqa_relaxed_overall,0.6376,0.009615793331418735
+Baseline,18000,docvqa_val_anls,0.6775884603912571,0.005972234236435759
+Baseline,18000,infovqa_val_anls,0.27154318420389256,0.007164903131667027
+Baseline,18000,mme_total_score,1336.922769107643,
+Baseline,18000,mmmu_val_mmmu_acc,0.28667,
+Baseline,18000,mmstar_average,0.34482796716566916,
+Baseline,18000,ocrbench_ocrbench_accuracy,0.533,
+Baseline,18000,seedbench_seed_all,0.5543079488604781,
+Baseline,18000,textvqa_val_exact_match,0.5666399999999999,0.006713392287599574
+Baseline,19000,ai2d_exact_match,0.4682642487046632,0.008981008686994101
+Baseline,19000,average,0.4899006713916878,
+Baseline,19000,average_rank,2.7,
+Baseline,19000,chartqa_relaxed_overall,0.6444,0.009575809858898698
+Baseline,19000,docvqa_val_anls,0.678226526479947,0.005970619221588814
+Baseline,19000,infovqa_val_anls,0.26993847247278,0.0071348470764911525
+Baseline,19000,mme_total_score,1406.6628651460583,
+Baseline,19000,mmmu_val_mmmu_acc,0.28333,
+Baseline,19000,mmstar_average,0.356220913822775,
+Baseline,19000,ocrbench_ocrbench_accuracy,0.577,
+Baseline,19000,seedbench_seed_all,0.554585881045025,
+Baseline,19000,textvqa_val_exact_match,0.57714,0.0066918487914812905
+Baseline,20000,ai2d_exact_match,0.47571243523316065,0.00898853090258662
+Baseline,20000,average,0.4873169067639118,
+Baseline,20000,average_rank,3.1,
+Baseline,20000,chartqa_relaxed_overall,0.6336,0.009638338810708618
+Baseline,20000,docvqa_val_anls,0.6895214454380043,0.005896462073053767
+Baseline,20000,infovqa_val_anls,0.2655657550458317,0.007033265532032538
+Baseline,20000,mme_total_score,1324.6738695478193,
+Baseline,20000,mmmu_val_mmmu_acc,0.30111,
+Baseline,20000,mmstar_average,0.33806766134497995,
+Baseline,20000,ocrbench_ocrbench_accuracy,0.555,
+Baseline,20000,seedbench_seed_all,0.5587548638132296,
+Baseline,20000,textvqa_val_exact_match,0.56852,0.006720151338087659
+≥2,1000,ai2d_exact_match,0.2645725388601036,0.007939149662089442
+≥2,1000,average,0.2722931646460497,
+≥2,1000,average_rank,3.2,
+≥2,1000,chartqa_relaxed_overall,0.3664,0.009638338810708616
+≥2,1000,docvqa_val_anls,0.35825461497275807,0.005864292098743202
+≥2,1000,infovqa_val_anls,0.16722293767954274,0.0061333612650745235
+≥2,1000,mme_total_score,994.9906962785115,
+≥2,1000,mmmu_val_mmmu_acc,0.25111,
+≥2,1000,mmstar_average,0.2099224814637991,
+≥2,1000,ocrbench_ocrbench_accuracy,0.304,
+≥2,1000,seedbench_seed_all,0.24463590883824346,
+≥2,1000,textvqa_val_exact_match,0.28452000000000005,0.006179555914647949
+≥2,2000,ai2d_exact_match,0.2648963730569948,0.007942257693619753
+≥2,2000,average,0.3161289250086133,
+≥2,2000,average_rank,3.2,
+≥2,2000,chartqa_relaxed_overall,0.4476,0.00994692276581072
+≥2,2000,docvqa_val_anls,0.44553207035528164,0.006176982458046509
+≥2,2000,infovqa_val_anls,0.19690312157526974,0.00648399793536667
+≥2,2000,mme_total_score,1054.2768107242898,
+≥2,2000,mmmu_val_mmmu_acc,0.24778,
+≥2,2000,mmstar_average,0.20779488571532076,
+≥2,2000,ocrbench_ocrbench_accuracy,0.383,
+≥2,2000,seedbench_seed_all,0.2529738743746526,
+≥2,2000,textvqa_val_exact_match,0.39868,0.006677826756815335
+≥2,3000,ai2d_exact_match,0.2697538860103627,0.007988222765138163
+≥2,3000,average,0.34461871112110076,
+≥2,3000,average_rank,3.7,
+≥2,3000,chartqa_relaxed_overall,0.502,0.010001920583875201
+≥2,3000,docvqa_val_anls,0.4943706505276063,0.006276617082627261
+≥2,3000,infovqa_val_anls,0.21287605644341218,0.006682253709215569
+≥2,3000,mme_total_score,1162.2701080432173,
+≥2,3000,mmmu_val_mmmu_acc,0.25556,
+��2,3000,mmstar_average,0.21427247636922603,
+≥2,3000,ocrbench_ocrbench_accuracy,0.449,
+≥2,3000,seedbench_seed_all,0.2715953307392996,
+≥2,3000,textvqa_val_exact_match,0.43213999999999997,0.006742795943777913
+≥2,4000,ai2d_exact_match,0.27525906735751293,0.008038849490577975
+≥2,4000,average,0.37379440715652495,
+≥2,4000,average_rank,2.6,
+≥2,4000,chartqa_relaxed_overall,0.5356,0.009976616117083942
+≥2,4000,docvqa_val_anls,0.5415736777563739,0.006259488230977563
+≥2,4000,infovqa_val_anls,0.22392444384387208,0.00676041701311943
+≥2,4000,mme_total_score,1195.5438175270108,
+≥2,4000,mmmu_val_mmmu_acc,0.26667,
+≥2,4000,mmstar_average,0.2507897461569136,
+≥2,4000,ocrbench_ocrbench_accuracy,0.462,
+≥2,4000,seedbench_seed_all,0.34291272929405225,
+≥2,4000,textvqa_val_exact_match,0.46542,0.0067795602517745565
+≥2,5000,ai2d_exact_match,0.31055699481865284,0.008328207321163279
+≥2,5000,average,0.39445964778826137,
+≥2,5000,average_rank,3.3,
+≥2,5000,chartqa_relaxed_overall,0.552,0.00994776272300849
+≥2,5000,docvqa_val_anls,0.5556927230238289,0.006299299461817651
+≥2,5000,infovqa_val_anls,0.24261245038285142,0.007075738778751112
+≥2,5000,mme_total_score,1220.672168867547,
+≥2,5000,mmmu_val_mmmu_acc,0.27444,
+≥2,5000,mmstar_average,0.2522926162881413,
+≥2,5000,ocrbench_ocrbench_accuracy,0.467,
+≥2,5000,seedbench_seed_all,0.42768204558087825,
+≥2,5000,textvqa_val_exact_match,0.46785999999999994,0.006777889939974511
+≥2,6000,ai2d_exact_match,0.3325777202072539,0.008479663360791275
+≥2,6000,average,0.4101998600043759,
+≥2,6000,average_rank,3.7,
+≥2,6000,chartqa_relaxed_overall,0.5672,0.009911254067113462
+≥2,6000,docvqa_val_anls,0.5702012141050906,0.006263916894054504
+≥2,6000,infovqa_val_anls,0.21632587505016104,0.006473865748732477
+≥2,6000,mme_total_score,1313.7047819127652,
+≥2,6000,mmmu_val_mmmu_acc,0.28,
+≥2,6000,mmstar_average,0.28566177948176935,
+≥2,6000,ocrbench_ocrbench_accuracy,0.486,
+≥2,6000,seedbench_seed_all,0.4698721511951084,
+≥2,6000,textvqa_val_exact_match,0.48396000000000006,0.006801425994533192
+≥2,7000,ai2d_exact_match,0.35200777202072536,0.00859592682822483
+≥2,7000,average,0.4204955224344633,
+≥2,7000,average_rank,4.0,
+≥2,7000,chartqa_relaxed_overall,0.5712,0.00990007214980924
+≥2,7000,docvqa_val_anls,0.5850734578344774,0.006202520219850679
+≥2,7000,infovqa_val_anls,0.23449023638527144,0.0067906990453115955
+≥2,7000,mme_total_score,1247.423969587835,
+≥2,7000,mmmu_val_mmmu_acc,0.28444,
+≥2,7000,mmstar_average,0.29053864145068503,
+≥2,7000,ocrbench_ocrbench_accuracy,0.487,
+≥2,7000,seedbench_seed_all,0.48526959421901056,
+≥2,7000,textvqa_val_exact_match,0.49444000000000005,0.006796105847537853
+≥2,8000,ai2d_exact_match,0.3746761658031088,0.008711886524907496
+≥2,8000,average,0.43663916832315425,
+≥2,8000,average_rank,2.9,
+≥2,8000,chartqa_relaxed_overall,0.5816,0.00986790384075991
+≥2,8000,docvqa_val_anls,0.6028798426362394,0.006214872354058686
+≥2,8000,infovqa_val_anls,0.2535281850303886,0.0070045473889607445
+≥2,8000,mme_total_score,1300.5965386154462,
+≥2,8000,mmmu_val_mmmu_acc,0.27333,
+≥2,8000,mmstar_average,0.310944925107356,
+≥2,8000,ocrbench_ocrbench_accuracy,0.516,
+≥2,8000,seedbench_seed_all,0.5041133963312951,
+≥2,8000,textvqa_val_exact_match,0.51268,0.006798079603627737
+≥2,9000,ai2d_exact_match,0.3795336787564767,0.00873405559083709
+≥2,9000,average,0.43759974296352216,
+≥2,9000,average_rank,2.9,
+≥2,9000,chartqa_relaxed_overall,0.5884,0.009844437067525526
+≥2,9000,docvqa_val_anls,0.6175894644110065,0.0061700253612544395
+≥2,9000,infovqa_val_anls,0.24471327484068725,0.006934982517240646
+≥2,9000,mme_total_score,1258.1754701880752,
+≥2,9000,mmmu_val_mmmu_acc,0.27,
+≥2,9000,mmstar_average,0.2988526527658083,
+≥2,9000,ocrbench_ocrbench_accuracy,0.514,
+≥2,9000,seedbench_seed_all,0.5155086158977209,
+≥2,9000,textvqa_val_exact_match,0.5098,0.00680062068405066
+≥2,10000,ai2d_exact_match,0.407059585492228,0.008842319527489083
+≥2,10000,average,0.45127176699921406,
+≥2,10000,average_rank,3.1,
+≥2,10000,chartqa_relaxed_overall,0.5956,0.009817474681589429
+≥2,10000,docvqa_val_anls,0.6286443353240219,0.006128441640319587
+≥2,10000,infovqa_val_anls,0.25277210900180563,0.007055702724548255
+≥2,10000,mme_total_score,1320.1028411364546,
+≥2,10000,mmmu_val_mmmu_acc,0.27556,
+≥2,10000,mmstar_average,0.3429750538307907,
+≥2,10000,ocrbench_ocrbench_accuracy,0.523,
+≥2,10000,seedbench_seed_all,0.51467481934408,
+≥2,10000,textvqa_val_exact_match,0.5211600000000001,0.006783601870014644
+≥2,11000,ai2d_exact_match,0.41580310880829013,0.008870644443998564
+≥2,11000,average,0.4525862975952584,
+≥2,11000,average_rank,3.5,
+≥2,11000,chartqa_relaxed_overall,0.598,0.009808000752013664
+≥2,11000,docvqa_val_anls,0.6307438129106796,0.006133911991297053
+≥2,11000,infovqa_val_anls,0.25390014221903434,0.007050537280004977
+≥2,11000,mme_total_score,1302.5287114845937,
+≥2,11000,mmmu_val_mmmu_acc,0.29333,
+≥2,11000,mmstar_average,0.303972877343168,
+���2,11000,ocrbench_ocrbench_accuracy,0.523,
+≥2,11000,seedbench_seed_all,0.5281267370761534,
+≥2,11000,textvqa_val_exact_match,0.5264,0.006786826961404041
+≥2,12000,ai2d_exact_match,0.43426165803108807,0.008921034830887027
+≥2,12000,average,0.46342874141175217,
+≥2,12000,average_rank,2.7,
+≥2,12000,chartqa_relaxed_overall,0.6188,0.009715574144248037
+≥2,12000,docvqa_val_anls,0.6419729722202083,0.006094582531110984
+≥2,12000,infovqa_val_anls,0.24776952598966778,0.006784112219881613
+≥2,12000,mme_total_score,1255.4957983193276,
+≥2,12000,mmmu_val_mmmu_acc,0.27111,
+≥2,12000,mmstar_average,0.3424608032908198,
+≥2,12000,ocrbench_ocrbench_accuracy,0.541,
+≥2,12000,seedbench_seed_all,0.5306837131739855,
+≥2,12000,textvqa_val_exact_match,0.5428,0.006758192556691964
+≥2,13000,ai2d_exact_match,0.42843264248704666,0.008906491762178375
+≥2,13000,average,0.4611120038339278,
+≥2,13000,average_rank,3.8,
+≥2,13000,chartqa_relaxed_overall,0.606,0.00977465178546074
+≥2,13000,docvqa_val_anls,0.6433656711922792,0.0061086851054902285
+≥2,13000,infovqa_val_anls,0.2535479547381062,0.006989226376396767
+≥2,13000,mme_total_score,1360.003101240496,
+≥2,13000,mmmu_val_mmmu_acc,0.28556,
+≥2,13000,mmstar_average,0.3320394092229932,
+≥2,13000,ocrbench_ocrbench_accuracy,0.526,
+≥2,13000,seedbench_seed_all,0.5362423568649249,
+≥2,13000,textvqa_val_exact_match,0.53882,0.006765393974568386
+≥2,14000,ai2d_exact_match,0.44689119170984454,0.008948245073044956
+≥2,14000,average,0.47130833654714216,
+≥2,14000,average_rank,2.8,
+≥2,14000,chartqa_relaxed_overall,0.6216,0.009701702181065136
+≥2,14000,docvqa_val_anls,0.6619108814388047,0.006015398975274413
+≥2,14000,infovqa_val_anls,0.2567040650730957,0.006986745571340195
+≥2,14000,mme_total_score,1310.3628451380553,
+≥2,14000,mmmu_val_mmmu_acc,0.28333,
+≥2,14000,mmstar_average,0.3315916867003111,
+≥2,14000,ocrbench_ocrbench_accuracy,0.547,
+≥2,14000,seedbench_seed_all,0.5409672040022234,
+≥2,14000,textvqa_val_exact_match,0.55178,0.006748546131944198
+≥2,15000,ai2d_exact_match,0.4523963730569948,0.00895827521082005
+≥2,15000,average,0.4720211465604895,
+≥2,15000,average_rank,3.5,
+≥2,15000,chartqa_relaxed_overall,0.62,0.009709671008043154
+≥2,15000,docvqa_val_anls,0.6679183447758706,0.005982903367170995
+≥2,15000,infovqa_val_anls,0.24815705436683513,0.006864270716284432
+≥2,15000,mme_total_score,1236.2534013605443,
+≥2,15000,mmmu_val_mmmu_acc,0.29889,
+≥2,15000,mmstar_average,0.3351456007635487,
+≥2,15000,ocrbench_ocrbench_accuracy,0.527,
+≥2,15000,seedbench_seed_all,0.5453029460811561,
+≥2,15000,textvqa_val_exact_match,0.55338,0.006735012041373013
+≥2,16000,ai2d_exact_match,0.44624352331606215,0.008946992176353898
+≥2,16000,average,0.4766960932538844,
+≥2,16000,average_rank,3.2,
+≥2,16000,chartqa_relaxed_overall,0.612,0.009747841205275417
+≥2,16000,docvqa_val_anls,0.6754589054855508,0.005966817690473989
+≥2,16000,infovqa_val_anls,0.27323519213464514,0.007206289716945655
+≥2,16000,mme_total_score,1305.906762705082,
+≥2,16000,mmmu_val_mmmu_acc,0.29,
+≥2,16000,mmstar_average,0.34328884147265926,
+≥2,16000,ocrbench_ocrbench_accuracy,0.555,
+≥2,16000,seedbench_seed_all,0.5410783768760422,
+≥2,16000,textvqa_val_exact_match,0.55396,0.00674076785464787
+≥2,17000,ai2d_exact_match,0.4485103626943005,0.008951310133709686
+≥2,17000,average,0.4803744475549501,
+≥2,17000,average_rank,3.3,
+≥2,17000,chartqa_relaxed_overall,0.6352,0.009629406741314642
+≥2,17000,docvqa_val_anls,0.6735387256928971,0.006001868055856522
+≥2,17000,infovqa_val_anls,0.2713449738427,0.007231154690666275
+≥2,17000,mme_total_score,1302.8314325730291,
+≥2,17000,mmmu_val_mmmu_acc,0.28667,
+≥2,17000,mmstar_average,0.33631999578132954,
+≥2,17000,ocrbench_ocrbench_accuracy,0.571,
+≥2,17000,seedbench_seed_all,0.542745969983324,
+≥2,17000,textvqa_val_exact_match,0.5580400000000001,0.006741465801458199
+≥2,18000,ai2d_exact_match,0.46113989637305697,0.008971933568013592
+≥2,18000,average,0.48745721111983964,
+≥2,18000,average_rank,2.6,
+≥2,18000,chartqa_relaxed_overall,0.6276,0.009670817229291067
+≥2,18000,docvqa_val_anls,0.6812777947859573,0.005935773909547658
+≥2,18000,infovqa_val_anls,0.27095882924867687,0.007164605404977649
+≥2,18000,mme_total_score,1289.7513005202081,
+≥2,18000,mmmu_val_mmmu_acc,0.31556,
+≥2,18000,mmstar_average,0.35401030852022664,
+≥2,18000,ocrbench_ocrbench_accuracy,0.564,
+≥2,18000,seedbench_seed_all,0.5505280711506393,
+≥2,18000,textvqa_val_exact_match,0.5620400000000001,0.00673487040527694
+≥2,19000,ai2d_exact_match,0.4698834196891192,0.008982814668850815
+≥2,19000,average,0.48664836716175586,
+≥2,19000,average_rank,3.2,
+≥2,19000,chartqa_relaxed_overall,0.6276,0.009670817229291067
+≥2,19000,docvqa_val_anls,0.6838077764263535,0.005944136929785695
+≥2,19000,infovqa_val_anls,0.26757170067350106,0.007096398035000058
+≥2,19000,mme_total_score,1310.4946978791518,
+≥2,19000,mmmu_val_mmmu_acc,0.29444,
+≥2,19000,mmstar_average,0.365800601107629,
+≥2,19000,ocrbench_ocrbench_accuracy,0.559,
+≥2,19000,seedbench_seed_all,0.5532518065591996,
+≥2,19000,textvqa_val_exact_match,0.55848,0.006735717623117797
+≥2,20000,ai2d_exact_match,0.4727979274611399,0.008985826352357515
+≥2,20000,average,0.4887875980209429,
+≥2,20000,average_rank,3.4,
+≥2,20000,chartqa_relaxed_overall,0.6392,0.00960657371300514
+≥2,20000,docvqa_val_anls,0.6828620051596259,0.005923332769971399
+≥2,20000,infovqa_val_anls,0.2701274975234547,0.007055868134029247
+≥2,20000,mme_total_score,1323.9108643457382,
+≥2,20000,mmmu_val_mmmu_acc,0.30222,
+≥2,20000,mmstar_average,0.33931189145504953,
+≥2,20000,ocrbench_ocrbench_accuracy,0.57,
+≥2,20000,seedbench_seed_all,0.5563090605892163,
+≥2,20000,textvqa_val_exact_match,0.56626,0.0067178082936069205
+≥3,1000,ai2d_exact_match,0.2691062176165803,0.007982164708643914
+≥3,1000,average,0.27573784261835144,
+≥3,1000,average_rank,2.8,
+≥3,1000,chartqa_relaxed_overall,0.352,0.009553790345406665
+≥3,1000,docvqa_val_anls,0.3425840937939014,0.005755186508181206
+≥3,1000,infovqa_val_anls,0.1714752271538445,0.006218691549786442
+≥3,1000,mme_total_score,1013.1872749099639,
+≥3,1000,mmmu_val_mmmu_acc,0.24778,
+≥3,1000,mmstar_average,0.2075589805205699,
+≥3,1000,ocrbench_ocrbench_accuracy,0.324,
+≥3,1000,seedbench_seed_all,0.24891606448026682,
+≥3,1000,textvqa_val_exact_match,0.31822,0.006368399926474836
+≥3,2000,ai2d_exact_match,0.25647668393782386,0.007859644922870102
+≥3,2000,average,0.32059377128504934,
+≥3,2000,average_rank,2.8,
+≥3,2000,chartqa_relaxed_overall,0.4628,0.009974279848861338
+≥3,2000,docvqa_val_anls,0.4518369496978485,0.00619300217721929
+≥3,2000,infovqa_val_anls,0.21204013425009277,0.006820894774458214
+≥3,2000,mme_total_score,1118.8858543417368,
+≥3,2000,mmmu_val_mmmu_acc,0.25222,
+≥3,2000,mmstar_average,0.20454842826555975,
+≥3,2000,ocrbench_ocrbench_accuracy,0.376,
+≥3,2000,seedbench_seed_all,0.25514174541411894,
+≥3,2000,textvqa_val_exact_match,0.41428000000000004,0.006714956027174666
+≥3,3000,ai2d_exact_match,0.25259067357512954,0.007820231277456426
+≥3,3000,average,0.35341646277484595,
+≥3,3000,average_rank,2.4,
+≥3,3000,chartqa_relaxed_overall,0.5208,0.00999334232158103
+≥3,3000,docvqa_val_anls,0.49758866181984457,0.00626460182861003
+≥3,3000,infovqa_val_anls,0.21333414080666746,0.0067509043256437935
+≥3,3000,mme_total_score,1165.3744497799119,
+≥3,3000,mmmu_val_mmmu_acc,0.26,
+≥3,3000,mmstar_average,0.2652435492500152,
+≥3,3000,ocrbench_ocrbench_accuracy,0.442,
+≥3,3000,seedbench_seed_all,0.29205113952195666,
+≥3,3000,textvqa_val_exact_match,0.43714,0.006763850531672249
+≥3,4000,ai2d_exact_match,0.28303108808290156,0.008107723290508887
+≥3,4000,average,0.37496255619498237,
+≥3,4000,average_rank,3.4,
+≥3,4000,chartqa_relaxed_overall,0.5412,0.009967987174315731
+≥3,4000,docvqa_val_anls,0.5296261512617491,0.006274192303767133
+≥3,4000,infovqa_val_anls,0.2050381576936679,0.006416570814061769
+≥3,4000,mme_total_score,1119.7681072428973,
+≥3,4000,mmmu_val_mmmu_acc,0.25556,
+≥3,4000,mmstar_average,0.24897141082880767,
+≥3,4000,ocrbench_ocrbench_accuracy,0.47,
+≥3,4000,seedbench_seed_all,0.3811561978877154,
+≥3,4000,textvqa_val_exact_match,0.46007999999999993,0.006793769924125808
+≥3,5000,ai2d_exact_match,0.3248056994818653,0.008428647470081763
+≥3,5000,average,0.3977887563101667,
+≥3,5000,average_rank,2.8,
+≥3,5000,chartqa_relaxed_overall,0.5544,0.009942625323290008
+≥3,5000,docvqa_val_anls,0.553669449701632,0.006282439058750721
+≥3,5000,infovqa_val_anls,0.20821650889148954,0.006430552192683275
+≥3,5000,mme_total_score,1326.9777911164465,
+≥3,5000,mmmu_val_mmmu_acc,0.26444,
+≥3,5000,mmstar_average,0.279759822424129,
+≥3,5000,ocrbench_ocrbench_accuracy,0.487,
+≥3,5000,seedbench_seed_all,0.43718732629238466,
+≥3,5000,textvqa_val_exact_match,0.47062,0.0067917147023207275
+≥3,6000,ai2d_exact_match,0.3536269430051813,0.008604903043803527
+≥3,6000,average,0.41524300122458385,
+≥3,6000,average_rank,3.1,
+≥3,6000,chartqa_relaxed_overall,0.568,0.009909070383761948
+≥3,6000,docvqa_val_anls,0.5722640243712676,0.00625854154899254
+≥3,6000,infovqa_val_anls,0.2204869348964998,0.00662088578415522
+≥3,6000,mme_total_score,1270.3575430172068,
+≥3,6000,mmmu_val_mmmu_acc,0.26556,
+≥3,6000,mmstar_average,0.2958896090262379,
+≥3,6000,ocrbench_ocrbench_accuracy,0.497,
+≥3,6000,seedbench_seed_all,0.47909949972206783,
+≥3,6000,textvqa_val_exact_match,0.48526,0.006795924028171543
+≥3,7000,ai2d_exact_match,0.3805051813471503,0.00873837769131663
+≥3,7000,average,0.42920372592352884,
+≥3,7000,average_rank,2.7,
+≥3,7000,chartqa_relaxed_overall,0.5728,0.009895414680177737
+≥3,7000,docvqa_val_anls,0.5922749765517075,0.006249497802747461
+≥3,7000,infovqa_val_anls,0.23025261139769496,0.006777932440928761
+≥3,7000,mme_total_score,1289.3664465786314,
+≥3,7000,mmmu_val_mmmu_acc,0.27111,
+≥3,7000,mmstar_average,0.3153601470057574,
+≥3,7000,ocrbench_ocrbench_accuracy,0.498,
+≥3,7000,seedbench_seed_all,0.4991106170094497,
+≥3,7000,textvqa_val_exact_match,0.50342,0.006801949281110862
+≥3,8000,ai2d_exact_match,0.39799222797927464,0.008809880751131852
+≥3,8000,average,0.438180751977588,
+≥3,8000,average_rank,2.4,
+≥3,8000,chartqa_relaxed_overall,0.5844,0.009858475126140203
+≥3,8000,docvqa_val_anls,0.6044755547364623,0.006202062618138765
+≥3,8000,infovqa_val_anls,0.21693088745597935,0.006529416377309533
+≥3,8000,mme_total_score,1187.3639455782313,
+≥3,8000,mmmu_val_mmmu_acc,0.28667,
+≥3,8000,mmstar_average,0.31735843114519735,
+≥3,8000,ocrbench_ocrbench_accuracy,0.506,
+≥3,8000,seedbench_seed_all,0.5193996664813786,
+≥3,8000,textvqa_val_exact_match,0.5104,0.0067972647853171315
+≥3,9000,ai2d_exact_match,0.407059585492228,0.008842319527489083
+≥3,9000,average,0.44395606448032265,
+≥3,9000,average_rank,3.0,
+≥3,9000,chartqa_relaxed_overall,0.598,0.009808000752013664
+≥3,9000,docvqa_val_anls,0.6107522318987826,0.006184930065074595
+≥3,9000,infovqa_val_anls,0.2347778400526839,0.0067525186273140235
+≥3,9000,mme_total_score,1195.0110044017606,
+≥3,9000,mmmu_val_mmmu_acc,0.28222,
+≥3,9000,mmstar_average,0.3264280968647572,
+≥3,9000,ocrbench_ocrbench_accuracy,0.521,
+≥3,9000,seedbench_seed_all,0.5162868260144525,
+≥3,9000,textvqa_val_exact_match,0.4990799999999999,0.00679372222366579
+≥3,10000,ai2d_exact_match,0.41580310880829013,0.008870644443998564
+≥3,10000,average,0.4524021135685592,
+≥3,10000,average_rank,2.5,
+≥3,10000,chartqa_relaxed_overall,0.5992,0.00980317218424473
+≥3,10000,docvqa_val_anls,0.6291907180725226,0.0061343676879221844
+≥3,10000,infovqa_val_anls,0.2282836442456148,0.006711844883510513
+≥3,10000,mme_total_score,1326.8972589035614,
+≥3,10000,mmmu_val_mmmu_acc,0.30111,
+≥3,10000,mmstar_average,0.3402582102457474,
+≥3,10000,ocrbench_ocrbench_accuracy,0.522,
+≥3,10000,seedbench_seed_all,0.5240133407448583,
+≥3,10000,textvqa_val_exact_match,0.51176,0.006789754092169055
+≥3,11000,ai2d_exact_match,0.42389896373056996,0.008894308540753343
+≥3,11000,average,0.45530296075039445,
+≥3,11000,average_rank,3.3,
+≥3,11000,chartqa_relaxed_overall,0.5992,0.00980317218424473
+≥3,11000,docvqa_val_anls,0.637004884118944,0.0060952660672868655
+≥3,11000,infovqa_val_anls,0.24182483065748125,0.006800414154487266
+≥3,11000,mme_total_score,1229.9441776710685,
+≥3,11000,mmmu_val_mmmu_acc,0.28556,
+≥3,11000,mmstar_average,0.3210406141609519,
+≥3,11000,ocrbench_ocrbench_accuracy,0.532,
+≥3,11000,seedbench_seed_all,0.5272373540856031,
+≥3,11000,textvqa_val_exact_match,0.52996,0.006774485841130848
+≥3,12000,ai2d_exact_match,0.4378238341968912,0.008929303814062614
+≥3,12000,average,0.4603808175211579,
+≥3,12000,average_rank,2.8,
+≥3,12000,chartqa_relaxed_overall,0.6036,0.009784943231599163
+≥3,12000,docvqa_val_anls,0.6425836471445318,0.006082856374953106
+≥3,12000,infovqa_val_anls,0.23921346499497054,0.00674373988949671
+≥3,12000,mme_total_score,1253.4613845538215,
+≥3,12000,mmmu_val_mmmu_acc,0.28,
+≥3,12000,mmstar_average,0.3402058443723711,
+≥3,12000,ocrbench_ocrbench_accuracy,0.533,
+≥3,12000,seedbench_seed_all,0.5370205669816565,
+≥3,12000,textvqa_val_exact_match,0.52998,0.006788538632972067
+≥3,13000,ai2d_exact_match,0.4410621761658031,0.0089364152923413
+≥3,13000,average,0.46617815773624777,
+≥3,13000,average_rank,2.8,
+≥3,13000,chartqa_relaxed_overall,0.6116,0.009749676839741497
+≥3,13000,docvqa_val_anls,0.6435913615068958,0.006093449845266186
+≥3,13000,infovqa_val_anls,0.24655403627027533,0.0068431739840280935
+≥3,13000,mme_total_score,1338.6154461784713,
+≥3,13000,mmmu_val_mmmu_acc,0.29556,
+≥3,13000,mmstar_average,0.33746561777886447,
+≥3,13000,ocrbench_ocrbench_accuracy,0.543,
+≥3,13000,seedbench_seed_all,0.5384102279043913,
+≥3,13000,textvqa_val_exact_match,0.5383600000000001,0.006773985492742893
+≥3,14000,ai2d_exact_match,0.4426813471502591,0.008939826412531762
+≥3,14000,average,0.46514162030247774,
+≥3,14000,average_rank,3.6,
+≥3,14000,chartqa_relaxed_overall,0.6104,0.009755142291143075
+≥3,14000,docvqa_val_anls,0.6522898002805984,0.006013616663077038
+≥3,14000,infovqa_val_anls,0.23824160343368236,0.006685403314320424
+≥3,14000,mme_total_score,1290.797318927571,
+≥3,14000,mmmu_val_mmmu_acc,0.29111,
+≥3,14000,mmstar_average,0.34665083130189556,
+≥3,14000,ocrbench_ocrbench_accuracy,0.533,
+≥3,14000,seedbench_seed_all,0.5418010005558643,
+≥3,14000,textvqa_val_exact_match,0.5300999999999999,0.006785072250248203
+≥3,15000,ai2d_exact_match,0.4536917098445596,0.00896047438220532
+≥3,15000,average,0.47760694744777243,
+≥3,15000,average_rank,2.9,
+≥3,15000,chartqa_relaxed_overall,0.612,0.009747841205275417
+≥3,15000,docvqa_val_anls,0.6656498012528964,0.006035466987037702
+≥3,15000,infovqa_val_anls,0.2625991131461808,0.007063916588796129
+≥3,15000,mme_total_score,1285.4465786314527,
+≥3,15000,mmmu_val_mmmu_acc,0.30222,
+≥3,15000,mmstar_average,0.3502185231309507,
+≥3,15000,ocrbench_ocrbench_accuracy,0.558,
+≥3,15000,seedbench_seed_all,0.5500833796553641,
+≥3,15000,textvqa_val_exact_match,0.544,0.0067575389652278954
+≥3,16000,ai2d_exact_match,0.4689119170984456,0.008981742470016596
+≥3,16000,average,0.4804309718902879,
+≥3,16000,average_rank,2.5,
+≥3,16000,chartqa_relaxed_overall,0.6204,0.009707689307588963
+≥3,16000,docvqa_val_anls,0.6742164965149466,0.0059800657435710326
+≥3,16000,infovqa_val_anls,0.2633355771988975,0.00704601997176055
+≥3,16000,mme_total_score,1288.4584833933575,
+≥3,16000,mmmu_val_mmmu_acc,0.29556,
+≥3,16000,mmstar_average,0.3443487528651147,
+≥3,16000,ocrbench_ocrbench_accuracy,0.55,
+≥3,16000,seedbench_seed_all,0.5508060033351863,
+≥3,16000,textvqa_val_exact_match,0.5563,0.006742548063668376
+≥3,17000,ai2d_exact_match,0.45595854922279794,0.008964175733819342
+≥3,17000,average,0.4809373657329622,
+≥3,17000,average_rank,3.3,
+≥3,17000,chartqa_relaxed_overall,0.6204,0.009707689307588963
+≥3,17000,docvqa_val_anls,0.6739488016448908,0.005975889304414765
+≥3,17000,infovqa_val_anls,0.2580649809644441,0.007031141926644411
+≥3,17000,mme_total_score,1230.4375750300119,
+≥3,17000,mmmu_val_mmmu_acc,0.29444,
+≥3,17000,mmstar_average,0.3444925534276732,
+≥3,17000,ocrbench_ocrbench_accuracy,0.578,
+≥3,17000,seedbench_seed_all,0.5565314063368538,
+≥3,17000,textvqa_val_exact_match,0.5466,0.006752985159298985
+≥3,18000,ai2d_exact_match,0.45919689119170987,0.008969138793675545
+≥3,18000,average,0.48088758936650067,
+≥3,18000,average_rank,3.5,
+≥3,18000,chartqa_relaxed_overall,0.6252,0.009683361554563506
+≥3,18000,docvqa_val_anls,0.675384499731014,0.005997750609265588
+≥3,18000,infovqa_val_anls,0.2579974692510198,0.0070128299378415275
+≥3,18000,mme_total_score,1234.843237294918,
+≥3,18000,mmmu_val_mmmu_acc,0.3,
+≥3,18000,mmstar_average,0.3363850750308216,
+≥3,18000,ocrbench_ocrbench_accuracy,0.566,
+≥3,18000,seedbench_seed_all,0.5558643690939411,
+≥3,18000,textvqa_val_exact_match,0.55196,0.006755291146330729
+≥3,19000,ai2d_exact_match,0.4634067357512953,0.008975020819363737
+≥3,19000,average,0.4861360634692545,
+≥3,19000,average_rank,3.3,
+≥3,19000,chartqa_relaxed_overall,0.6312,0.009651522406019766
+≥3,19000,docvqa_val_anls,0.6819220996842664,0.005927423649467908
+≥3,19000,infovqa_val_anls,0.26277439983326806,0.007102707331042042
+≥3,19000,mme_total_score,1337.9653861544616,
+≥3,19000,mmmu_val_mmmu_acc,0.29889,
+≥3,19000,mmstar_average,0.34778832316957964,
+≥3,19000,ocrbench_ocrbench_accuracy,0.574,
+≥3,19000,seedbench_seed_all,0.5614230127848805,
+≥3,19000,textvqa_val_exact_match,0.55382,0.006743039020727005
+≥3,20000,ai2d_exact_match,0.4841321243523316,0.008994621193008031
+≥3,20000,average,0.4916087790351852,
+≥3,20000,average_rank,2.3,
+≥3,20000,chartqa_relaxed_overall,0.638,0.009613499245701268
+≥3,20000,docvqa_val_anls,0.6839168937073106,0.005936410873687919
+≥3,20000,infovqa_val_anls,0.25441216838205727,0.006890877173562315
+≥3,20000,mme_total_score,1330.3037214885953,
+≥3,20000,mmmu_val_mmmu_acc,0.31,
+≥3,20000,mmstar_average,0.35052721898280503,
+≥3,20000,ocrbench_ocrbench_accuracy,0.572,
+≥3,20000,seedbench_seed_all,0.5630906058921623,
+≥3,20000,textvqa_val_exact_match,0.5684000000000001,0.00672360984783302
+≥4,1000,ai2d_exact_match,0.26360103626943004,0.00792979255467583
+≥4,1000,average,0.26922373369534647,
+≥4,1000,average_rank,3.3,
+≥4,1000,chartqa_relaxed_overall,0.3488,0.009533718094861256
+≥4,1000,docvqa_val_anls,0.3599045480096881,0.005885735735631119
+≥4,1000,infovqa_val_anls,0.17148252623256244,0.0061724612150041895
+≥4,1000,mme_total_score,1104.3533413365346,
+≥4,1000,mmmu_val_mmmu_acc,0.24,
+≥4,1000,mmstar_average,0.21109041770474804,
+≥4,1000,ocrbench_ocrbench_accuracy,0.29,
+≥4,1000,seedbench_seed_all,0.24313507504168982,
+≥4,1000,textvqa_val_exact_match,0.295,0.006241441429527609
+≥4,2000,ai2d_exact_match,0.26295336787564766,0.007923526907377255
+≥4,2000,average,0.3192621308996215,
+≥4,2000,average_rank,2.9,
+≥4,2000,chartqa_relaxed_overall,0.4644,0.009976616117083942
+≥4,2000,docvqa_val_anls,0.44610634422212336,0.006125661837378556
+≥4,2000,infovqa_val_anls,0.19012118870963063,0.006420072608935975
+≥4,2000,mme_total_score,1052.5613245298118,
+≥4,2000,mmmu_val_mmmu_acc,0.23889,
+≥4,2000,mmstar_average,0.22088912220303317,
+≥4,2000,ocrbench_ocrbench_accuracy,0.389,
+≥4,2000,seedbench_seed_all,0.262479155086159,
+≥4,2000,textvqa_val_exact_match,0.39852,0.006693677836929181
+≥4,3000,ai2d_exact_match,0.2545336787564767,0.007840040862810524
+≥4,3000,average,0.34899183633853254,
+≥4,3000,average_rank,3.1,
+≥4,3000,chartqa_relaxed_overall,0.5144,0.009997851710018818
+≥4,3000,docvqa_val_anls,0.5122633926443586,0.006250170224123374
+≥4,3000,infovqa_val_anls,0.21839296983497156,0.006786311152255019
+≥4,3000,mme_total_score,1148.9873949579833,
+≥4,3000,mmmu_val_mmmu_acc,0.24667,
+≥4,3000,mmstar_average,0.23884910949080793,
+≥4,3000,ocrbench_ocrbench_accuracy,0.426,
+≥4,3000,seedbench_seed_all,0.29927737632017787,
+≥4,3000,textvqa_val_exact_match,0.43054,0.006761938068430401
+≥4,4000,ai2d_exact_match,0.2814119170984456,0.00809362228799086
+≥4,4000,average,0.3808723899304912,
+≥4,4000,average_rank,2.3,
+≥4,4000,chartqa_relaxed_overall,0.536,0.009976041728231964
+≥4,4000,docvqa_val_anls,0.5444976153718191,0.006262351643342788
+≥4,4000,infovqa_val_anls,0.22943118895386538,0.006865542219383826
+≥4,4000,mme_total_score,1161.4330732292917,
+≥4,4000,mmmu_val_mmmu_acc,0.26889,
+≥4,4000,mmstar_average,0.2546319608241094,
+≥4,4000,ocrbench_ocrbench_accuracy,0.459,
+≥4,4000,seedbench_seed_all,0.39988882712618123,
+≥4,4000,textvqa_val_exact_match,0.4541,0.006780990662644609
+≥4,5000,ai2d_exact_match,0.31573834196891193,0.00836578020190971
+≥4,5000,average,0.4004212057382194,
+≥4,5000,average_rank,2.7,
+≥4,5000,chartqa_relaxed_overall,0.5544,0.009942625323290008
+≥4,5000,docvqa_val_anls,0.556855142418819,0.006267140081468451
+≥4,5000,infovqa_val_anls,0.23435340618373432,0.006883129487757931
+≥4,5000,mme_total_score,1145.157863145258,
+≥4,5000,mmmu_val_mmmu_acc,0.26556,
+≥4,5000,mmstar_average,0.2888277743020811,
+≥4,5000,ocrbench_ocrbench_accuracy,0.475,
+≥4,5000,seedbench_seed_all,0.445136186770428,
+≥4,5000,textvqa_val_exact_match,0.46792,0.0067973094238147356
+≥4,6000,ai2d_exact_match,0.38471502590673573,0.008756678690415541
+≥4,6000,average,0.42131977921781544,
+≥4,6000,average_rank,3.0,
+≥4,6000,chartqa_relaxed_overall,0.556,0.00993907007952043
+≥4,6000,docvqa_val_anls,0.5727106862384739,0.006269180765398416
+≥4,6000,infovqa_val_anls,0.2310709838980833,0.006744459748098398
+≥4,6000,mme_total_score,1139.8311324529811,
+≥4,6000,mmmu_val_mmmu_acc,0.27,
+≥4,6000,mmstar_average,0.30779610290926424,
+≥4,6000,ocrbench_ocrbench_accuracy,0.492,
+≥4,6000,seedbench_seed_all,0.4933852140077821,
+≥4,6000,textvqa_val_exact_match,0.4841999999999999,0.006796772117869219
+≥4,7000,ai2d_exact_match,0.39281088082901555,0.008789930274160654
+≥4,7000,average,0.42891500537341953,
+≥4,7000,average_rank,2.9,
+≥4,7000,chartqa_relaxed_overall,0.576,0.009885782289560632
+≥4,7000,docvqa_val_anls,0.5907488324071782,0.006231156163373406
+≥4,7000,infovqa_val_anls,0.24013816441297325,0.006930097636315065
+≥4,7000,mme_total_score,1162.137755102041,
+≥4,7000,mmmu_val_mmmu_acc,0.27556,
+≥4,7000,mmstar_average,0.29752599783778977,
+≥4,7000,ocrbench_ocrbench_accuracy,0.504,
+≥4,7000,seedbench_seed_all,0.5001111728738188,
+≥4,7000,textvqa_val_exact_match,0.48333999999999994,0.006805450147517214
+≥4,8000,ai2d_exact_match,0.4164507772020725,0.008872627955954676
+≥4,8000,average,0.43574351275219425,
+≥4,8000,average_rank,3.5,
+≥4,8000,chartqa_relaxed_overall,0.5808,0.009870537726284339
+≥4,8000,docvqa_val_anls,0.6057226019616091,0.0061946427553956785
+≥4,8000,infovqa_val_anls,0.2476713069705094,0.006953489019987495
+≥4,8000,mme_total_score,1170.280612244898,
+≥4,8000,mmmu_val_mmmu_acc,0.26778,
+≥4,8000,mmstar_average,0.30520454953605713,
+≥4,8000,ocrbench_ocrbench_accuracy,0.496,
+≥4,8000,seedbench_seed_all,0.5082823790994997,
+≥4,8000,textvqa_val_exact_match,0.49378,0.006806491606223952
+≥4,9000,ai2d_exact_match,0.42357512953367876,0.008893409023558714
+≥4,9000,average,0.441337144868937,
+≥4,9000,average_rank,3.4,
+≥4,9000,chartqa_relaxed_overall,0.578,0.00987954665846924
+≥4,9000,docvqa_val_anls,0.6243353881540346,0.006123815047404004
+≥4,9000,infovqa_val_anls,0.2437398253282973,0.00692277294272151
+≥4,9000,mme_total_score,1255.001700680272,
+≥4,9000,mmmu_val_mmmu_acc,0.26778,
+≥4,9000,mmstar_average,0.31167080905344985,
+≥4,9000,ocrbench_ocrbench_accuracy,0.512,
+≥4,9000,seedbench_seed_all,0.5116731517509727,
+≥4,9000,textvqa_val_exact_match,0.49926000000000004,0.006799642454386958
+≥4,10000,ai2d_exact_match,0.44462435233160624,0.00894379269709736
+≥4,10000,average,0.4536388119594498,
+≥4,10000,average_rank,3.6,
+≥4,10000,chartqa_relaxed_overall,0.5992,0.00980317218424473
+≥4,10000,docvqa_val_anls,0.6264595846035441,0.006147505656275056
+≥4,10000,infovqa_val_anls,0.2598110483089896,0.00706252458320144
+≥4,10000,mme_total_score,1192.952080832333,
+≥4,10000,mmmu_val_mmmu_acc,0.28556,
+≥4,10000,mmstar_average,0.3186673407344323,
+≥4,10000,ocrbench_ocrbench_accuracy,0.52,
+≥4,10000,seedbench_seed_all,0.5205669816564759,
+≥4,10000,textvqa_val_exact_match,0.5078600000000001,0.006802447996107573
+≥4,11000,ai2d_exact_match,0.4536917098445596,0.008960474382205324
+≥4,11000,average,0.46152725733636885,
+≥4,11000,average_rank,2.3,
+≥4,11000,chartqa_relaxed_overall,0.6004,0.009798282427824488
+≥4,11000,docvqa_val_anls,0.6401993666501584,0.0060898160255800525
+≥4,11000,infovqa_val_anls,0.2552761209118603,0.007046581941151624
+≥4,11000,mme_total_score,1246.6340536214486,
+≥4,11000,mmmu_val_mmmu_acc,0.28,
+≥4,11000,mmstar_average,0.3347344054467562,
+≥4,11000,ocrbench_ocrbench_accuracy,0.533,
+≥4,11000,seedbench_seed_all,0.5306837131739855,
+≥4,11000,textvqa_val_exact_match,0.5257599999999999,0.006772980077619183
+≥4,12000,ai2d_exact_match,0.45919689119170987,0.008969138793675545
+≥4,12000,average,0.46100484234717837,
+≥4,12000,average_rank,3.7,
+≥4,12000,chartqa_relaxed_overall,0.5956,0.009817474681589429
+≥4,12000,docvqa_val_anls,0.6409268162702743,0.006097072959583667
+≥4,12000,infovqa_val_anls,0.26230050824120466,0.007170670588017343
+≥4,12000,mme_total_score,1168.390556222489,
+≥4,12000,mmmu_val_mmmu_acc,0.26889,
+≥4,12000,mmstar_average,0.3293674421306994,
+≥4,12000,ocrbench_ocrbench_accuracy,0.538,
+≥4,12000,seedbench_seed_all,0.5314619232907171,
+≥4,12000,textvqa_val_exact_match,0.5233,0.006791483405661084
+≥4,13000,ai2d_exact_match,0.46081606217616583,0.008971477299154906
+≥4,13000,average,0.46897968661537504,
+≥4,13000,average_rank,3.0,
+≥4,13000,chartqa_relaxed_overall,0.6084,0.00976411343463736
+≥4,13000,docvqa_val_anls,0.6557097904355208,0.006045284321472833
+≥4,13000,infovqa_val_anls,0.25716935409374025,0.007037968981507592
+≥4,13000,mme_total_score,1214.4760904361744,
+≥4,13000,mmmu_val_mmmu_acc,0.27444,
+≥4,13000,mmstar_average,0.35062705343328215,
+≥4,13000,ocrbench_ocrbench_accuracy,0.542,
+≥4,13000,seedbench_seed_all,0.5388549193996665,
+≥4,13000,textvqa_val_exact_match,0.5328,0.006772208248489718
+≥4,14000,ai2d_exact_match,0.4637305699481865,0.008975446629055962
+≥4,14000,average,0.46882712562329804,
+≥4,14000,average_rank,3.4,
+≥4,14000,chartqa_relaxed_overall,0.6052,0.009778109662477129
+≥4,14000,docvqa_val_anls,0.6600293980607723,0.006003818486747537
+≥4,14000,infovqa_val_anls,0.2604896578960276,0.0070806001081496605
+≥4,14000,mme_total_score,1180.2360944377751,
+≥4,14000,mmmu_val_mmmu_acc,0.29889,
+≥4,14000,mmstar_average,0.3370405135985262,
+≥4,14000,ocrbench_ocrbench_accuracy,0.532,
+≥4,14000,seedbench_seed_all,0.5311839911061701,
+≥4,14000,textvqa_val_exact_match,0.53088,0.006765681045393848
+≥4,15000,ai2d_exact_match,0.469559585492228,0.008982461065390123
+≥4,15000,average,0.47678210727691706,
+≥4,15000,average_rank,3.0,
+≥4,15000,chartqa_relaxed_overall,0.6228,0.009695651925812239
+≥4,15000,docvqa_val_anls,0.668732849209273,0.006002172541493102
+≥4,15000,infovqa_val_anls,0.2541377129865746,0.006911037097155498
+≥4,15000,mme_total_score,1198.8395358143257,
+≥4,15000,mmmu_val_mmmu_acc,0.28111,
+≥4,15000,mmstar_average,0.3574887121899482,
+≥4,15000,ocrbench_ocrbench_accuracy,0.558,
+≥4,15000,seedbench_seed_all,0.5421901056142301,
+≥4,15000,textvqa_val_exact_match,0.53702,0.0067620891069120025
+≥4,16000,ai2d_exact_match,0.4689119170984456,0.00898174247001659
+≥4,16000,average,0.47623501147363423,
+≥4,16000,average_rank,3.7,
+≥4,16000,chartqa_relaxed_overall,0.6184,0.009717527882093043
+≥4,16000,docvqa_val_anls,0.664711612332228,0.006033753206179003
+≥4,16000,infovqa_val_anls,0.26137627968800997,0.0069587136315641595
+≥4,16000,mme_total_score,1223.327631052421,
+≥4,16000,mmmu_val_mmmu_acc,0.27778,
+≥4,16000,mmstar_average,0.3532221312757646,
+≥4,16000,ocrbench_ocrbench_accuracy,0.545,
+≥4,16000,seedbench_seed_all,0.5476931628682602,
+≥4,16000,textvqa_val_exact_match,0.54902,0.006730591957147508
+≥4,17000,ai2d_exact_match,0.47830310880829013,0.008990677331728418
+≥4,17000,average,0.4815150623543914,
+≥4,17000,average_rank,2.8,
+≥4,17000,chartqa_relaxed_overall,0.6208,0.009705700605814084
+≥4,17000,docvqa_val_anls,0.6784945768946954,0.005958779114256312
+≥4,17000,infovqa_val_anls,0.27415576971914574,0.007211057524316044
+≥4,17000,mme_total_score,1267.6510604241696,
+≥4,17000,mmmu_val_mmmu_acc,0.27889,
+≥4,17000,mmstar_average,0.35485659715149337,
+≥4,17000,ocrbench_ocrbench_accuracy,0.55,
+≥4,17000,seedbench_seed_all,0.5479155086158978,
+≥4,17000,textvqa_val_exact_match,0.5502199999999999,0.006738803500215962
+≥4,18000,ai2d_exact_match,0.4795984455958549,0.008991659681159872
+≥4,18000,average,0.4839656525796875,
+≥4,18000,average_rank,3.1,
+≥4,18000,chartqa_relaxed_overall,0.6228,0.009695651925812239
+≥4,18000,docvqa_val_anls,0.680615041882376,0.005957029786047422
+≥4,18000,infovqa_val_anls,0.27507992619170296,0.007267921800589956
+≥4,18000,mme_total_score,1226.5048019207684,
+≥4,18000,mmmu_val_mmmu_acc,0.28111,
+≥4,18000,mmstar_average,0.35607565298805366,
+≥4,18000,ocrbench_ocrbench_accuracy,0.555,
+≥4,18000,seedbench_seed_all,0.5532518065591996,
+≥4,18000,textvqa_val_exact_match,0.55216,0.006730239676654988
+≥4,19000,ai2d_exact_match,0.4734455958549223,0.008986453895645547
+≥4,19000,average,0.485443851233213,
+≥4,19000,average_rank,3.0,
+≥4,19000,chartqa_relaxed_overall,0.6276,0.009670817229291067
+≥4,19000,docvqa_val_anls,0.690884348495626,0.005908240141234498
+≥4,19000,infovqa_val_anls,0.2676836840845966,0.007165567282387595
+≥4,19000,mme_total_score,1323.2516006402561,
+≥4,19000,mmmu_val_mmmu_acc,0.28556,
+≥4,19000,mmstar_average,0.33406913716627346,
+≥4,19000,ocrbench_ocrbench_accuracy,0.584,
+≥4,19000,seedbench_seed_all,0.5414118954974986,
+≥4,19000,textvqa_val_exact_match,0.56434,0.006692191716171407
+≥4,20000,ai2d_exact_match,0.4876943005181347,0.008996428218289526
+≥4,20000,average,0.4906341423361293,
+≥4,20000,average_rank,3.3,
+≥4,20000,chartqa_relaxed_overall,0.6284,0.009666579183001631
+≥4,20000,docvqa_val_anls,0.6887236251150223,0.005918556723502163
+≥4,20000,infovqa_val_anls,0.2809124119459898,0.007354611102020885
+≥4,20000,mme_total_score,1254.5532212885155,
+≥4,20000,mmmu_val_mmmu_acc,0.29333,
+≥4,20000,mmstar_average,0.34736535367392096,
+≥4,20000,ocrbench_ocrbench_accuracy,0.572,
+≥4,20000,seedbench_seed_all,0.5508615897720957,
+≥4,20000,textvqa_val_exact_match,0.56642,0.00672606309106159
+≥5,1000,ai2d_exact_match,0.26327720207253885,0.007926662492947056
+≥5,1000,average,0.27709006947371073,
+≥5,1000,average_rank,2.6,
+≥5,1000,chartqa_relaxed_overall,0.3412,0.009484144853461517
+≥5,1000,docvqa_val_anls,0.36296241117667905,0.005852839558467308
+≥5,1000,infovqa_val_anls,0.17994878830754762,0.006336933369747534
+≥5,1000,mme_total_score,968.375450180072,
+≥5,1000,mmmu_val_mmmu_acc,0.26667,
+≥5,1000,mmstar_average,0.22684359669162246,
+≥5,1000,ocrbench_ocrbench_accuracy,0.301,
+≥5,1000,seedbench_seed_all,0.25152862701500833,
+≥5,1000,textvqa_val_exact_match,0.30038,0.006282823083071704
+≥5,2000,ai2d_exact_match,0.27331606217616583,0.008021157484423327
+≥5,2000,average,0.318491261297989,
+≥5,2000,average_rank,3.2,
+≥5,2000,chartqa_relaxed_overall,0.4524,0.009956573172519544
+≥5,2000,docvqa_val_anls,0.4578740641673747,0.006180081722767688
+≥5,2000,infovqa_val_anls,0.1919057230410833,0.006401757863597739
+≥5,2000,mme_total_score,1031.2603041216487,
+≥5,2000,mmmu_val_mmmu_acc,0.24667,
+≥5,2000,mmstar_average,0.21129996032951712,
+≥5,2000,ocrbench_ocrbench_accuracy,0.383,
+≥5,2000,seedbench_seed_all,0.25597554196775985,
+≥5,2000,textvqa_val_exact_match,0.39398,0.0066750028503822015
+≥5,3000,ai2d_exact_match,0.2661917098445596,0.007954634970279373
+≥5,3000,average,0.3470898411915701,
+≥5,3000,average_rank,3.1,
+≥5,3000,chartqa_relaxed_overall,0.4888,0.009999490983443667
+≥5,3000,docvqa_val_anls,0.5063663265388635,0.006269377896147078
+≥5,3000,infovqa_val_anls,0.2002412084672373,0.006449644926640854
+≥5,3000,mme_total_score,1176.8578431372548,
+≥5,3000,mmmu_val_mmmu_acc,0.25889,
+≥5,3000,mmstar_average,0.2226891646728035,
+≥5,3000,ocrbench_ocrbench_accuracy,0.422,
+≥5,3000,seedbench_seed_all,0.32229016120066706,
+≥5,3000,textvqa_val_exact_match,0.43633999999999995,0.006743513614961789
+≥5,4000,ai2d_exact_match,0.32091968911917096,0.008402150106895235
+≥5,4000,average,0.38454946481840957,
+≥5,4000,average_rank,3.0,
+≥5,4000,chartqa_relaxed_overall,0.5244,0.009990083919101193
+≥5,4000,docvqa_val_anls,0.5408182220870532,0.0062304604635426315
+≥5,4000,infovqa_val_anls,0.21034975209325477,0.006529781109938355
+≥5,4000,mme_total_score,1186.4263705482194,
+≥5,4000,mmmu_val_mmmu_acc,0.26556,
+≥5,4000,mmstar_average,0.26918979355147643,
+≥5,4000,ocrbench_ocrbench_accuracy,0.452,
+≥5,4000,seedbench_seed_all,0.4339077265147304,
+≥5,4000,textvqa_val_exact_match,0.4438,0.006776008770579609
+≥5,5000,ai2d_exact_match,0.3494170984455959,0.008581339503665948
+≥5,5000,average,0.4053929772745627,
+≥5,5000,average_rank,2.9,
+≥5,5000,chartqa_relaxed_overall,0.546,0.009959582185560013
+≥5,5000,docvqa_val_anls,0.5611769594797935,0.006252030837783964
+≥5,5000,infovqa_val_anls,0.2283202771889911,0.006874345513158979
+≥5,5000,mme_total_score,1179.6603641456581,
+≥5,5000,mmmu_val_mmmu_acc,0.27556,
+≥5,5000,mmstar_average,0.28276518409209217,
+≥5,5000,ocrbench_ocrbench_accuracy,0.464,
+≥5,5000,seedbench_seed_all,0.4750972762645914,
+≥5,5000,textvqa_val_exact_match,0.4662,0.0067984671677640855
+≥5,6000,ai2d_exact_match,0.3636658031088083,0.008658158841882573
+≥5,6000,average,0.41623598541202544,
+≥5,6000,average_rank,2.6,
+≥5,6000,chartqa_relaxed_overall,0.5584,0.009933541468098847
+≥5,6000,docvqa_val_anls,0.5839255211800125,0.006223251970774856
+≥5,6000,infovqa_val_anls,0.23899504944949723,0.007013133491201096
+≥5,6000,mme_total_score,1252.7314925970388,
+≥5,6000,mmmu_val_mmmu_acc,0.27222,
+≥5,6000,mmstar_average,0.3101670336024846,
+≥5,6000,ocrbench_ocrbench_accuracy,0.473,
+≥5,6000,seedbench_seed_all,0.49483046136742637,
+≥5,6000,textvqa_val_exact_match,0.45092,0.006772193384764505
+≥5,7000,ai2d_exact_match,0.42033678756476683,0.008884198538329093
+≥5,7000,average,0.4338560588435303,
+≥5,7000,average_rank,2.5,
+≥5,7000,chartqa_relaxed_overall,0.5692,0.00990574548014469
+≥5,7000,docvqa_val_anls,0.5924368390904757,0.006231022369252223
+≥5,7000,infovqa_val_anls,0.23945153983485024,0.007006534034576772
+≥5,7000,mme_total_score,1315.113445378151,
+≥5,7000,mmmu_val_mmmu_acc,0.3,
+≥5,7000,mmstar_average,0.31063340423564356,
+≥5,7000,ocrbench_ocrbench_accuracy,0.488,
+≥5,7000,seedbench_seed_all,0.5067259588660367,
+≥5,7000,textvqa_val_exact_match,0.47791999999999996,0.006793800546466833
+≥5,8000,ai2d_exact_match,0.42908031088082904,0.008908169846895226
+≥5,8000,average,0.43778255861533233,
+≥5,8000,average_rank,3.0,
+≥5,8000,chartqa_relaxed_overall,0.5752,0.009888230116554488
+≥5,8000,docvqa_val_anls,0.6032859006895523,0.006193925022795706
+≥5,8000,infovqa_val_anls,0.24493490021598546,0.007008771158507111
+≥5,8000,mme_total_score,1304.6824729891955,
+≥5,8000,mmmu_val_mmmu_acc,0.28667,
+≥5,8000,mmstar_average,0.31703546216629863,
+≥5,8000,ocrbench_ocrbench_accuracy,0.487,
+≥5,8000,seedbench_seed_all,0.5096164535853251,
+≥5,8000,textvqa_val_exact_match,0.48722,0.006804659800386776
+≥5,9000,ai2d_exact_match,0.42940414507772023,0.008909003051055709
+≥5,9000,average,0.44649777930382,
+≥5,9000,average_rank,2.5,
+≥5,9000,chartqa_relaxed_overall,0.5792,0.009875725592704212
+≥5,9000,docvqa_val_anls,0.6158422097964253,0.00617698110304048
+≥5,9000,infovqa_val_anls,0.24039717009699607,0.0068877247346275485
+≥5,9000,mme_total_score,1379.7254901960782,
+≥5,9000,mmmu_val_mmmu_acc,0.29889,
+≥5,9000,mmstar_average,0.3280690123874737,
+≥5,9000,ocrbench_ocrbench_accuracy,0.513,
+≥5,9000,seedbench_seed_all,0.5234574763757643,
+≥5,9000,textvqa_val_exact_match,0.4902200000000001,0.006801597067199211
+≥5,10000,ai2d_exact_match,0.45012953367875647,0.00895427929990258
+≥5,10000,average,0.4555663389491801,
+≥5,10000,average_rank,2.9,
+≥5,10000,chartqa_relaxed_overall,0.5844,0.009858475126140203
+≥5,10000,docvqa_val_anls,0.6189420793161403,0.006040465868816934
+≥5,10000,infovqa_val_anls,0.24850918819779613,0.007091394184737253
+≥5,10000,mme_total_score,1235.7704081632655,
+≥5,10000,mmmu_val_mmmu_acc,0.27667,
+≥5,10000,mmstar_average,0.34675895640940496,
+≥5,10000,ocrbench_ocrbench_accuracy,0.528,
+≥5,10000,seedbench_seed_all,0.5291272929405225,
+≥5,10000,textvqa_val_exact_match,0.51756,0.006786717998284417
+≥5,11000,ai2d_exact_match,0.4566062176165803,0.008965198879336196
+≥5,11000,average,0.458748059148796,
+≥5,11000,average_rank,2.9,
+≥5,11000,chartqa_relaxed_overall,0.5916,0.0098327233755248
+≥5,11000,docvqa_val_anls,0.633602507666147,0.006134122729213928
+≥5,11000,infovqa_val_anls,0.2621320066294427,0.007275786683175354
+≥5,11000,mme_total_score,1326.4276710684273,
+≥5,11000,mmmu_val_mmmu_acc,0.27667,
+≥5,11000,mmstar_average,0.34479339575773355,
+≥5,11000,ocrbench_ocrbench_accuracy,0.517,
+≥5,11000,seedbench_seed_all,0.5311284046692607,
+≥5,11000,textvqa_val_exact_match,0.5152000000000001,0.006786619456012555
+≥5,12000,ai2d_exact_match,0.45919689119170987,0.008969138793675545
+≥5,12000,average,0.4644995480385772,
+≥5,12000,average_rank,2.3,
+≥5,12000,chartqa_relaxed_overall,0.596,0.009815912634917984
+≥5,12000,docvqa_val_anls,0.6453485539631237,0.006065269954977215
+≥5,12000,infovqa_val_anls,0.2685572578806166,0.007278550841020009
+≥5,12000,mme_total_score,1374.9406762705082,
+≥5,12000,mmmu_val_mmmu_acc,0.28444,
+≥5,12000,mmstar_average,0.35205377405882654,
+≥5,12000,ocrbench_ocrbench_accuracy,0.519,
+≥5,12000,seedbench_seed_all,0.5350194552529183,
+≥5,12000,textvqa_val_exact_match,0.52088,0.006777757204160069
+≥5,13000,ai2d_exact_match,0.4640544041450777,0.008975868633841907
+≥5,13000,average,0.4696984757423332,
+≥5,13000,average_rank,2.7,
+≥5,13000,chartqa_relaxed_overall,0.608,0.00976588700628918
+≥5,13000,docvqa_val_anls,0.6599237778239753,0.006035894149838363
+≥5,13000,infovqa_val_anls,0.25759117282312316,0.007107246020667877
+≥5,13000,mme_total_score,1326.0453181272508,
+≥5,13000,mmmu_val_mmmu_acc,0.28667,
+≥5,13000,mmstar_average,0.35252858336464304,
+≥5,13000,ocrbench_ocrbench_accuracy,0.533,
+≥5,13000,seedbench_seed_all,0.5330183435241801,
+≥5,13000,textvqa_val_exact_match,0.5325,0.006770636476998357
+≥5,14000,ai2d_exact_match,0.4689119170984456,0.00898174247001659
+≥5,14000,average,0.47293227498131896,
+≥5,14000,average_rank,2.7,
+≥5,14000,chartqa_relaxed_overall,0.614,0.009738559226822298
+≥5,14000,docvqa_val_anls,0.6583491716485876,0.0060256160547597325
+≥5,14000,infovqa_val_anls,0.26613522559599984,0.0071532088405842145
+≥5,14000,mme_total_score,1278.5425170068027,
+≥5,14000,mmmu_val_mmmu_acc,0.28,
+≥5,14000,mmstar_average,0.35624004153386235,
+≥5,14000,ocrbench_ocrbench_accuracy,0.55,
+≥5,14000,seedbench_seed_all,0.5454141189549749,
+≥5,14000,textvqa_val_exact_match,0.5173399999999999,0.006787096420087393
+≥5,15000,ai2d_exact_match,0.4740932642487047,0.008987066275159846
+≥5,15000,average,0.47568039073709784,
+≥5,15000,average_rank,3.0,
+≥5,15000,chartqa_relaxed_overall,0.602,0.00979166741164548
+≥5,15000,docvqa_val_anls,0.6649825816931088,0.006012202194059076
+≥5,15000,infovqa_val_anls,0.2659187859072639,0.007233849219121225
+≥5,15000,mme_total_score,1301.498799519808,
+≥5,15000,mmmu_val_mmmu_acc,0.30333,
+≥5,15000,mmstar_average,0.363574304462402,
+≥5,15000,ocrbench_ocrbench_accuracy,0.536,
+≥5,15000,seedbench_seed_all,0.5402445803224013,
+≥5,15000,textvqa_val_exact_match,0.53098,0.006774896882281907
+≥5,16000,ai2d_exact_match,0.47538860103626945,0.008988245555188545
+≥5,16000,average,0.48103362013771567,
+≥5,16000,average_rank,2.6,
+≥5,16000,chartqa_relaxed_overall,0.6172,0.009723347231923635
+≥5,16000,docvqa_val_anls,0.6661394800733964,0.006000339067695713
+≥5,16000,infovqa_val_anls,0.27200681388207976,0.007361243845813883
+≥5,16000,mme_total_score,1312.4185674269709,
+≥5,16000,mmmu_val_mmmu_acc,0.30667,
+≥5,16000,mmstar_average,0.352673072573432,
+≥5,16000,ocrbench_ocrbench_accuracy,0.553,
+≥5,16000,seedbench_seed_all,0.5483046136742635,
+≥5,16000,textvqa_val_exact_match,0.53792,0.0067618902203356104
+≥5,17000,ai2d_exact_match,0.4740932642487047,0.008987066275159846
+≥5,17000,average,0.4842246444979549,
+≥5,17000,average_rank,3.1,
+≥5,17000,chartqa_relaxed_overall,0.6252,0.009683361554563506
+≥5,17000,docvqa_val_anls,0.6727784028551866,0.005982986502554192
+≥5,17000,infovqa_val_anls,0.273461783643309,0.00736211121641681
+≥5,17000,mme_total_score,1256.561224489796,
+≥5,17000,mmmu_val_mmmu_acc,0.31889,
+≥5,17000,mmstar_average,0.35664172938975786,
+≥5,17000,ocrbench_ocrbench_accuracy,0.545,
+≥5,17000,seedbench_seed_all,0.549916620344636,
+≥5,17000,textvqa_val_exact_match,0.5420400000000001,0.006760567190239792
+≥5,18000,ai2d_exact_match,0.4802461139896373,0.008992128148477658
+≥5,18000,average,0.48392876207158253,
+≥5,18000,average_rank,2.9,
+≥5,18000,chartqa_relaxed_overall,0.6252,0.009683361554563506
+≥5,18000,docvqa_val_anls,0.68034033548242,0.005889534044935538
+≥5,18000,infovqa_val_anls,0.28015930560506613,0.007380855727131182
+≥5,18000,mme_total_score,1380.5266106442577,
+≥5,18000,mmmu_val_mmmu_acc,0.3,
+≥5,18000,mmstar_average,0.34877922919246646,
+≥5,18000,ocrbench_ocrbench_accuracy,0.549,
+≥5,18000,seedbench_seed_all,0.5529738743746526,
+≥5,18000,textvqa_val_exact_match,0.5386599999999999,0.0067648675941745775
+≥5,19000,ai2d_exact_match,0.4805699481865285,0.008992356706334513
+≥5,19000,average,0.49271643602329757,
+≥5,19000,average_rank,2.8,
+≥5,19000,chartqa_relaxed_overall,0.6248,0.009685427559111736
+≥5,19000,docvqa_val_anls,0.6825000217737053,0.005934471601355602
+≥5,19000,infovqa_val_anls,0.2841253071402532,0.007403930662950274
+≥5,19000,mme_total_score,1261.751700680272,
+≥5,19000,mmmu_val_mmmu_acc,0.32,
+≥5,19000,mmstar_average,0.3611420745688909,
+≥5,19000,ocrbench_ocrbench_accuracy,0.572,
+≥5,19000,seedbench_seed_all,0.5550305725403002,
+≥5,19000,textvqa_val_exact_match,0.5542799999999999,0.006739897741383979
+≥5,20000,ai2d_exact_match,0.4844559585492228,0.008994804366753555
+≥5,20000,average,0.49543136618963995,
+≥5,20000,average_rank,2.9,
+≥5,20000,chartqa_relaxed_overall,0.638,0.009613499245701268
+≥5,20000,docvqa_val_anls,0.688451623661496,0.005905200575549553
+≥5,20000,infovqa_val_anls,0.2789607789162199,0.007289443671361681
+≥5,20000,mme_total_score,1296.043617446979,
+≥5,20000,mmmu_val_mmmu_acc,0.3,
+≥5,20000,mmstar_average,0.3804272197382418,
+≥5,20000,ocrbench_ocrbench_accuracy,0.577,
+≥5,20000,seedbench_seed_all,0.5560867148415787,
+≥5,20000,textvqa_val_exact_match,0.5555,0.006734970078953051

data/root_seq_write.json

@@ -0,0 +1,402 @@
+{
+  "metadata": {
+    "thread_counts": [
+      1,
+      4,
+      8,
+      16,
+      32,
+      64
+    ],
+    "io_size_labels": [
+      "64K",
+      "128K",
+      "256K",
+      "512K",
+      "1M",
+      "2M",
+      "4M",
+      "8M"
+    ]
+  },
+  "data": {
+    "CPUONLY": {
+      "throughput": [
+        [
+          0.043969,
+          0.062935,
+          0.080797,
+          0.157091,
+          0.3077,
+          0.598654,
+          1.084552,
+          1.084701
+        ],
+        [
+          0.044531,
+          0.062424,
+          0.080556,
+          0.156628,
+          0.30637,
+          0.608223,
+          1.08397,
+          1.084789
+        ],
+        [
+          0.043632,
+          0.062382,
+          0.080396,
+          0.156629,
+          0.306526,
+          0.729085,
+          1.084771,
+          1.085995
+        ],
+        [
+          0.043793,
+          0.062397,
+          0.080297,
+          0.156834,
+          0.306295,
+          0.756486,
+          1.085467,
+          1.085047
+        ],
+        [
+          0.043555,
+          0.062383,
+          0.080054,
+          0.156675,
+          0.30618,
+          0.766355,
+          1.086493,
+          1.080395
+        ],
+        [
+          0.043404,
+          0.062419,
+          0.080311,
+          0.155996,
+          0.30614,
+          0.739887,
+          1.074057,
+          1.060276
+        ]
+      ],
+      "iops": [
+        [
+          720.3934706968972,
+          515.5649962742374,
+          330.94474805670217,
+          321.72234399113466,
+          315.0851103645625,
+          306.5107571132344,
+          277.6454231293006,
+          138.84167942556795
+        ],
+        [
+          729.5955805913294,
+          511.3748990953006,
+          329.9557080359582,
+          320.7742159722403,
+          313.72299189760935,
+          311.4099328329025,
+          277.4963309052229,
+          138.85294975205963
+        ],
+        [
+          714.8665771507461,
+          511.0316195303459,
+          329.3035751188262,
+          320.7765456361971,
+          313.88300902990926,
+          373.2916044469156,
+          277.701256359883,
+          139.0073252907107
+        ],
+        [
+          717.4996330967786,
+          511.1582187372969,
+          328.8944792321785,
+          321.19553552373037,
+          313.6458418464666,
+          387.3207388162377,
+          277.87948460058607,
+          138.8859761116121
+        ],
+        [
+          713.6017449976126,
+          511.04259999550817,
+          327.89985520265907,
+          320.87121492689715,
+          313.5286114336872,
+          392.3740142861152,
+          278.1420891636299,
+          138.29058396481292
+        ],
+        [
+          711.1305463817035,
+          511.33953756002455,
+          328.9520454971924,
+          319.4807848220429,
+          313.4874718217851,
+          378.822372375107,
+          274.9585353694654,
+          135.71526956685778
+        ]
+      ]
+    },
+    "CPU_GPU": {
+      "throughput": [
+        [
+          0.043266,
+          0.061039,
+          0.077598,
+          0.144688,
+          0.265439,
+          0.46755,
+          0.745195,
+          0.984786
+        ],
+        [
+          0.043438,
+          0.062161,
+          0.080758,
+          0.155765,
+          0.306934,
+          0.650931,
+          1.087331,
+          1.087943
+        ],
+        [
+          0.043341,
+          0.062266,
+          0.080014,
+          0.156475,
+          0.306248,
+          0.734435,
+          1.087265,
+          1.07865
+        ],
+        [
+          0.043479,
+          0.062183,
+          0.080686,
+          0.156148,
+          0.305127,
+          0.742459,
+          1.077986,
+          1.080356
+        ],
+        [
+          0.04339,
+          0.061944,
+          0.079974,
+          0.155954,
+          0.304866,
+          0.735717,
+          1.080031,
+          1.078628
+        ],
+        [
+          0.043007,
+          0.061784,
+          0.079486,
+          0.155377,
+          0.3033,
+          0.746351,
+          1.077426,
+          1.065708
+        ]
+      ],
+      "iops": [
+        [
+          708.8687039870763,
+          500.0310015214285,
+          317.8411675135234,
+          296.32166762862266,
+          271.80969858555085,
+          239.3856667396793,
+          190.76999810852197,
+          126.05258230498497
+        ],
+        [
+          711.6827145606015,
+          509.21997244554143,
+          330.7839891429897,
+          319.00664604553435,
+          314.30070398746926,
+          333.2768364370973,
+          278.35685956641606,
+          139.2566758905968
+        ],
+        [
+          710.0932996519502,
+          510.0817354457645,
+          327.7371359652991,
+          320.4610929135827,
+          313.5975168083104,
+          376.03081326841266,
+          278.339869378998,
+          138.06718349148696
+        ],
+        [
+          712.3635674403308,
+          509.40610821891374,
+          330.4898387411662,
+          319.79135463802794,
+          312.4496135008945,
+          380.1391623804666,
+          275.9644748269969,
+          138.2856057348695
+        ],
+        [
+          710.909554597986,
+          507.4464188398417,
+          327.57547142580273,
+          319.3929428198887,
+          312.18309133073814,
+          376.6868527866089,
+          276.4880624703965,
+          138.06438127090303
+        ],
+        [
+          704.6327993263955,
+          506.1335940906205,
+          325.57576484222244,
+          318.2130870988752,
+          310.57871877710465,
+          382.13170328164813,
+          275.8209685059825,
+          136.41062338121034
+        ]
+      ]
+    },
+    "GPUD": {
+      "throughput": [
+        [
+          0.043112,
+          0.061729,
+          0.079366,
+          0.153501,
+          0.294881,
+          0.555678,
+          1.022165,
+          1.079885
+        ],
+        [
+          0.043293,
+          0.062371,
+          0.080618,
+          0.156828,
+          0.306714,
+          0.757807,
+          1.082543,
+          1.081335
+        ],
+        [
+          0.043216,
+          0.062372,
+          0.079963,
+          0.156,
+          0.30449,
+          0.733983,
+          1.083296,
+          1.081052
+        ],
+        [
+          0.043132,
+          0.06203,
+          0.080037,
+          0.155992,
+          0.305426,
+          0.734766,
+          1.083687,
+          1.078112
+        ],
+        [
+          0.043047,
+          0.062002,
+          0.079688,
+          0.155715,
+          0.305346,
+          0.733292,
+          1.082125,
+          1.076138
+        ],
+        [
+          0.042838,
+          0.06187,
+          0.080168,
+          0.155873,
+          0.304343,
+          0.737737,
+          1.068858,
+          1.061664
+        ]
+      ],
+      "iops": [
+        [
+          706.3538029102274,
+          505.6874644818375,
+          325.08227121166897,
+          314.369321362317,
+          301.9582938978947,
+          284.50701040485717,
+          261.67412039887546,
+          138.22522648779298
+        ],
+        [
+          709.307703386022,
+          510.947112518552,
+          330.21203337688104,
+          321.18443511834244,
+          314.07539704278963,
+          387.9972596331553,
+          277.1308860820274,
+          138.410903290691
+        ],
+        [
+          708.0580482881605,
+          510.95428892293927,
+          327.5290428742473,
+          319.48804226176173,
+          311.7972977422169,
+          375.7990856536964,
+          277.3236820505962,
+          138.3746079428659
+        ],
+        [
+          706.6713344687167,
+          508.15322009761513,
+          327.8335038767732,
+          319.47133538046825,
+          312.75609546837546,
+          376.20035032160007,
+          277.423791621676,
+          137.99830503659598
+        ],
+        [
+          705.2794305121502,
+          507.9194642362503,
+          326.4021505890313,
+          318.90515442102935,
+          312.67393051280527,
+          375.44546031443053,
+          277.02411969494,
+          137.7457048433716
+        ],
+        [
+          701.8540582079571,
+          506.8351859598479,
+          328.3664976340191,
+          319.2275496078547,
+          311.64681641178527,
+          377.72112019023353,
+          273.6276228003431,
+          135.89295166162492
+        ]
+      ]
+    }
+  }
+}

data/scratch_seq_write.json

@@ -0,0 +1,402 @@
+{
+  "metadata": {
+    "thread_counts": [
+      1,
+      4,
+      8,
+      16,
+      32,
+      64
+    ],
+    "io_size_labels": [
+      "64K",
+      "128K",
+      "256K",
+      "512K",
+      "1M",
+      "2M",
+      "4M",
+      "8M"
+    ]
+  },
+  "data": {
+    "CPUONLY": {
+      "throughput": [
+        [
+          1.201202,
+          1.690285,
+          2.358928,
+          2.824236,
+          5.288685,
+          9.284685,
+          14.505432,
+          12.239065
+        ],
+        [
+          4.420199,
+          6.509496,
+          9.95769,
+          10.742544,
+          13.288907,
+          22.720375,
+          26.5874,
+          26.312089
+        ],
+        [
+          8.273714,
+          12.148186,
+          13.707003,
+          14.896735,
+          17.914147,
+          24.581956,
+          24.371136,
+          25.559105
+        ],
+        [
+          14.319008,
+          19.398883,
+          20.152786,
+          20.108555,
+          22.97167,
+          23.62825,
+          25.387384,
+          24.905158
+        ],
+        [
+          18.274788,
+          23.250199,
+          24.567631,
+          24.831425,
+          25.524572,
+          26.406383,
+          26.036394,
+          25.144152
+        ],
+        [
+          20.591561,
+          24.287713,
+          21.549986,
+          24.279414,
+          26.594768,
+          23.333698,
+          26.383843,
+          26.354901
+        ]
+      ],
+      "iops": [
+        [
+          19680.498572902885,
+          13846.812301247413,
+          9662.1693302708,
+          5784.035148749556,
+          5415.613330084133,
+          4753.758634233075,
+          3713.3905331349897,
+          1566.6003165768102
+        ],
+        [
+          72420.53408753955,
+          53325.791528355425,
+          40786.69790096216,
+          22000.729420989486,
+          13607.840327705806,
+          11632.831797609619,
+          6806.374397955888,
+          3367.947444777335
+        ],
+        [
+          135556.52493993653,
+          99517.93763079736,
+          56143.88310299916,
+          30508.512949365882,
+          18344.08677196712,
+          12585.961346677055,
+          6239.010690350058,
+          3271.5654144586633
+        ],
+        [
+          234602.6263863431,
+          158915.6482055442,
+          82545.81064585822,
+          41182.32059300002,
+          23522.99003948308,
+          12097.664161237199,
+          6499.170214665355,
+          3187.860176161076
+        ],
+        [
+          299414.1321363972,
+          190465.62650239328,
+          100629.01600385213,
+          50854.758419018406,
+          26137.16157406106,
+          13520.068285400217,
+          6665.316826617809,
+          3218.4514807799937
+        ],
+        [
+          337372.13177364814,
+          198964.94414750757,
+          88268.74221033942,
+          49724.240364227066,
+          27233.04233154442,
+          11946.853274490775,
+          6754.263879056663,
+          3373.427387837718
+        ]
+      ]
+    },
+    "CPU_GPU": {
+      "throughput": [
+        [
+          0.640488,
+          0.885508,
+          1.168287,
+          1.157542,
+          1.504038,
+          1.748927,
+          1.903144,
+          1.955567
+        ],
+        [
+          1.962079,
+          2.409594,
+          2.176801,
+          2.160291,
+          2.293493,
+          2.162244,
+          2.055486,
+          2.015337
+        ],
+        [
+          2.313416,
+          2.516002,
+          2.156904,
+          2.225047,
+          2.137375,
+          2.032659,
+          2.065286,
+          2.071186
+        ],
+        [
+          2.167723,
+          2.264676,
+          2.139438,
+          2.045373,
+          2.048711,
+          2.058408,
+          2.106988,
+          1.996206
+        ],
+        [
+          2.281068,
+          2.239423,
+          2.1684,
+          2.434728,
+          2.084591,
+          2.052072,
+          2.021339,
+          6.720828
+        ],
+        [
+          4.75436,
+          6.453834,
+          6.91617,
+          7.06102,
+          6.644377,
+          2.107144,
+          1.976539,
+          1.853629
+        ]
+      ],
+      "iops": [
+        [
+          10493.75390269922,
+          7254.084119525284,
+          4785.3028021404825,
+          2370.6465746306044,
+          1540.135136941383,
+          895.4505515415741,
+          487.20483165648994,
+          250.3125930078555
+        ],
+        [
+          32146.701695687767,
+          19739.39773763156,
+          8916.176638528155,
+          4424.276616992854,
+          2348.53652670484,
+          1107.0690182692904,
+          526.2042984647344,
+          257.96312922440467
+        ],
+        [
+          37902.999634914915,
+          20611.09152507395,
+          8834.677940276075,
+          4556.895242738244,
+          2188.672372855926,
+          1040.7212663108983,
+          528.7132468731671,
+          265.11175252971276
+        ],
+        [
+          35515.97418734287,
+          18552.227837258546,
+          8763.13740967112,
+          4188.9232775454175,
+          2097.8804743814258,
+          1053.9047625296776,
+          539.3890505343694,
+          255.5143937941016
+        ],
+        [
+          37373.012155260534,
+          18345.35161349351,
+          8881.765587871974,
+          4986.3232026976,
+          2134.621418516056,
+          1050.6606307379845,
+          517.462750113315,
+          860.2660070924891
+        ],
+        [
+          77895.43185254499,
+          52869.80506248117,
+          28328.632644412693,
+          14460.96947876695,
+          6803.842311218816,
+          1078.8575218817803,
+          505.99403067626844,
+          237.26447400762223
+        ]
+      ]
+    },
+    "GPUD": {
+      "throughput": [
+        [
+          0.780713,
+          1.18524,
+          1.783481,
+          1.979619,
+          2.821537,
+          4.088575,
+          4.900446,
+          5.115048
+        ],
+        [
+          2.977402,
+          4.093661,
+          5.929054,
+          5.982685,
+          6.68657,
+          7.147719,
+          6.46833,
+          6.38473
+        ],
+        [
+          5.107965,
+          6.181268,
+          6.954231,
+          6.710225,
+          6.859411,
+          6.869249,
+          6.779233,
+          5.818478
+        ],
+        [
+          5.52386,
+          6.874029,
+          6.424485,
+          7.044277,
+          6.244672,
+          6.569815,
+          6.971813,
+          6.103482
+        ],
+        [
+          5.53895,
+          6.54437,
+          6.110592,
+          6.891137,
+          6.595009,
+          6.566134,
+          6.828472,
+          6.335114
+        ],
+        [
+          5.196324,
+          6.252264,
+          6.577701,
+          6.709694,
+          6.256264,
+          6.178591,
+          6.409033,
+          6.084137
+        ]
+      ],
+      "iops": [
+        [
+          12791.208711490808,
+          9709.488294836901,
+          7305.138279658066,
+          4054.2607156453464,
+          2889.253627273963,
+          2093.35028776338,
+          1254.5142417509228,
+          654.7260994283422
+        ],
+        [
+          48781.75590324929,
+          33535.268974505445,
+          24285.405397451606,
+          12252.538535713647,
+          6847.0476910851185,
+          3659.631981245416,
+          1655.8924540777045,
+          817.2454011683283
+        ],
+        [
+          83688.9062592321,
+          50636.943981751196,
+          28484.52916897171,
+          13742.541520380233,
+          7024.036947193166,
+          3517.0556096163305,
+          1735.4836281404082,
+          744.7652317618935
+        ],
+        [
+          90502.91701037118,
+          56312.04556870529,
+          26314.68889161115,
+          14426.67969713539,
+          6394.543972444186,
+          3363.7450981646302,
+          1784.784139905184,
+          781.2456472793523
+        ],
+        [
+          90750.15957519117,
+          53611.481296026286,
+          25028.98318567973,
+          14113.048284870958,
+          6753.288751882261,
+          3361.8605944698243,
+          1748.08872194917,
+          810.8946552041433
+        ],
+        [
+          85136.56730504583,
+          51218.54896584013,
+          26942.26269986482,
+          13741.453089601031,
+          6406.414544063209,
+          3163.438716053786,
+          1640.7124017017943,
+          778.769485853273
+        ]
+      ]
+    }
+  }
+}

data/typography_data.json

@@ -0,0 +1,2361 @@
+{
+  "metadata": {
+    "generated_at": "2025-09-22T16:11:54.157445",
+    "method": "umap_from_png_pixels",
+    "total_fonts": 293,
+    "umap_params": {
+      "n_neighbors": 15,
+      "min_dist": 1.0,
+      "n_components": 2,
+      "metric": "euclidean",
+      "random_state": 42
+    },
+    "data_source": "PNG pixel matrices (40x40)"
+  },
+  "fonts": [
+    {
+      "name": "Baumans",
+      "id": "baumans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baumans",
+      "x": 18.494117736816406,
+      "y": -0.46384990215301514
+    },
+    {
+      "name": "Castoro Titling",
+      "id": "castoro_titling",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Castoro+Titling",
+      "x": 19.08490562438965,
+      "y": -8.786286354064941
+    },
+    {
+      "name": "Almendra Sc",
+      "id": "almendra_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Almendra+Sc",
+      "x": 17.94493865966797,
+      "y": -8.496891975402832
+    },
+    {
+      "name": "Akaya Telivigala",
+      "id": "akaya_telivigala",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Akaya+Telivigala",
+      "x": 21.603944778442383,
+      "y": -5.4939398765563965
+    },
+    {
+      "name": "Afacad Flux",
+      "id": "afacad_flux",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Afacad+Flux",
+      "x": 17.90031623840332,
+      "y": -6.820926189422607
+    },
+    {
+      "name": "Carattere",
+      "id": "carattere",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Carattere",
+      "x": 16.19448471069336,
+      "y": -7.016509532928467
+    },
+    {
+      "name": "Abril Fatface",
+      "id": "abril_fatface",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Abril+Fatface",
+      "x": 20.192102432250977,
+      "y": -8.998359680175781
+    },
+    {
+      "name": "Asset",
+      "id": "asset",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Asset",
+      "x": 19.972240447998047,
+      "y": -1.2671222686767578
+    },
+    {
+      "name": "Be Vietnam Pro",
+      "id": "be_vietnam_pro",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Be+Vietnam+Pro",
+      "x": 23.365022659301758,
+      "y": -4.017947673797607
+    },
+    {
+      "name": "Alegreya Sans Sc",
+      "id": "alegreya_sans_sc",
+      "family": "serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alegreya+Sans+Sc",
+      "x": 20.48445701599121,
+      "y": -6.818283557891846
+    },
+    {
+      "name": "Aladin",
+      "id": "aladin",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Aladin",
+      "x": 21.421142578125,
+      "y": -4.155740737915039
+    },
+    {
+      "name": "Big Shoulders Inline",
+      "id": "big_shoulders_inline",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Big+Shoulders+Inline",
+      "x": 16.641128540039062,
+      "y": -4.189671516418457
+    },
+    {
+      "name": "Carter One",
+      "id": "carter_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Carter+One",
+      "x": 21.244213104248047,
+      "y": -2.1734020709991455
+    },
+    {
+      "name": "Alice",
+      "id": "alice",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alice",
+      "x": 18.316776275634766,
+      "y": -9.161249160766602
+    },
+    {
+      "name": "Amiri",
+      "id": "amiri",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Amiri",
+      "x": 17.296218872070312,
+      "y": -9.194451332092285
+    },
+    {
+      "name": "Arya",
+      "id": "arya",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arya",
+      "x": 20.68015480041504,
+      "y": -7.799196243286133
+    },
+    {
+      "name": "Andika",
+      "id": "andika",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Andika",
+      "x": 23.530542373657227,
+      "y": -4.775561332702637
+    },
+    {
+      "name": "Akatab",
+      "id": "akatab",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Akatab",
+      "x": 19.34527587890625,
+      "y": -6.275416374206543
+    },
+    {
+      "name": "Cantarell",
+      "id": "cantarell",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cantarell",
+      "x": 22.87999725341797,
+      "y": -6.737560749053955
+    },
+    {
+      "name": "Archivo Black",
+      "id": "archivo_black",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Archivo+Black",
+      "x": 22.111753463745117,
+      "y": -1.174342155456543
+    },
+    {
+      "name": "Barlow Condensed",
+      "id": "barlow_condensed",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Barlow+Condensed",
+      "x": 17.464378356933594,
+      "y": -4.241239547729492
+    },
+    {
+      "name": "Bree Serif",
+      "id": "bree_serif",
+      "family": "serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bree+Serif",
+      "x": 22.43787956237793,
+      "y": -2.9592504501342773
+    },
+    {
+      "name": "Actor",
+      "id": "actor",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Actor",
+      "x": 22.239173889160156,
+      "y": -5.835178375244141
+    },
+    {
+      "name": "Amiri Quran",
+      "id": "amiri_quran",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Amiri+Quran",
+      "x": 17.43636131286621,
+      "y": -9.012243270874023
+    },
+    {
+      "name": "Catamaran",
+      "id": "catamaran",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Catamaran",
+      "x": 22.96765899658203,
+      "y": -7.319466590881348
+    },
+    {
+      "name": "Cantata One",
+      "id": "cantata_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cantata+One",
+      "x": 21.11892318725586,
+      "y": -9.548401832580566
+    },
+    {
+      "name": "Anton",
+      "id": "anton",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anton",
+      "x": 18.142189025878906,
+      "y": -2.554518699645996
+    },
+    {
+      "name": "Bigshot One",
+      "id": "bigshot_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bigshot+One",
+      "x": 19.22067642211914,
+      "y": -8.094287872314453
+    },
+    {
+      "name": "Bitcount",
+      "id": "bitcount",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount",
+      "x": 17.018287658691406,
+      "y": -0.40733659267425537
+    },
+    {
+      "name": "Biorhyme Expanded",
+      "id": "biorhyme_expanded",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Biorhyme+Expanded",
+      "x": 19.0455322265625,
+      "y": -0.8312316536903381
+    },
+    {
+      "name": "Bonbon",
+      "id": "bonbon",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bonbon",
+      "x": 15.121182441711426,
+      "y": -5.263389587402344
+    },
+    {
+      "name": "Amethysta",
+      "id": "amethysta",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Amethysta",
+      "x": 21.001907348632812,
+      "y": -9.594971656799316
+    },
+    {
+      "name": "Arbutus",
+      "id": "arbutus",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arbutus",
+      "x": 21.7993221282959,
+      "y": -1.4154260158538818
+    },
+    {
+      "name": "Bowlby One",
+      "id": "bowlby_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bowlby+One",
+      "x": 20.656328201293945,
+      "y": -0.8706757426261902
+    },
+    {
+      "name": "Cambay",
+      "id": "cambay",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cambay",
+      "x": 21.44684410095215,
+      "y": -6.603434085845947
+    },
+    {
+      "name": "Audiowide",
+      "id": "audiowide",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Audiowide",
+      "x": 19.63788414001465,
+      "y": -1.476771354675293
+    },
+    {
+      "name": "Bitcount Prop Single",
+      "id": "bitcount_prop_single",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Prop+Single",
+      "x": 17.749900817871094,
+      "y": 0.07072576135396957
+    },
+    {
+      "name": "Blaka Hollow",
+      "id": "blaka_hollow",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Blaka+Hollow",
+      "x": 19.00209617614746,
+      "y": -3.3146474361419678
+    },
+    {
+      "name": "Anek Odia",
+      "id": "anek_odia",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Odia",
+      "x": 19.981218338012695,
+      "y": -6.394980430603027
+    },
+    {
+      "name": "Anek Latin",
+      "id": "anek_latin",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Latin",
+      "x": 19.357284545898438,
+      "y": -5.552064418792725
+    },
+    {
+      "name": "Baloo Tamma 2",
+      "id": "baloo_tamma_2",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baloo+Tamma+2",
+      "x": 22.396146774291992,
+      "y": -7.621818542480469
+    },
+    {
+      "name": "Butterfly Kids",
+      "id": "butterfly_kids",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Butterfly+Kids",
+      "x": 16.24427604675293,
+      "y": -6.357809543609619
+    },
+    {
+      "name": "Bhutuka Expanded One",
+      "id": "bhutuka_expanded_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bhutuka+Expanded+One",
+      "x": 17.062284469604492,
+      "y": -5.975480079650879
+    },
+    {
+      "name": "Abyssinica Sil",
+      "id": "abyssinica_sil",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Abyssinica+Sil",
+      "x": 20.37006950378418,
+      "y": -9.457380294799805
+    },
+    {
+      "name": "Blacknd White Picture",
+      "id": "blacknd_white_picture",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Blacknd+White+Picture",
+      "x": 17.998069763183594,
+      "y": -5.218323707580566
+    },
+    {
+      "name": "Archivo",
+      "id": "archivo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Archivo",
+      "x": 23.97217559814453,
+      "y": -3.5558722019195557
+    },
+    {
+      "name": "Anek Malayalam",
+      "id": "anek_malayalam",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Malayalam",
+      "x": 18.917207717895508,
+      "y": -6.025264263153076
+    },
+    {
+      "name": "Averia Serif Libre",
+      "id": "averia_serif_libre",
+      "family": "serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Averia+Serif+Libre",
+      "x": 20.519973754882812,
+      "y": -8.761312484741211
+    },
+    {
+      "name": "Alegreya",
+      "id": "alegreya",
+      "family": "serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alegreya",
+      "x": 17.29988670349121,
+      "y": -9.604442596435547
+    },
+    {
+      "name": "Cantora One",
+      "id": "cantora_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cantora+One",
+      "x": 22.678606033325195,
+      "y": -3.359496593475342
+    },
+    {
+      "name": "Ancizar Serif",
+      "id": "ancizar_serif",
+      "family": "serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Ancizar+Serif",
+      "x": 17.800979614257812,
+      "y": -8.884207725524902
+    },
+    {
+      "name": "Anuphan",
+      "id": "anuphan",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anuphan",
+      "x": 23.491052627563477,
+      "y": -6.040447235107422
+    },
+    {
+      "name": "Bungee Inline",
+      "id": "bungee_inline",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bungee+Inline",
+      "x": 20.4497127532959,
+      "y": -0.7007665038108826
+    },
+    {
+      "name": "Akaya Kanadaka",
+      "id": "akaya_kanadaka",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Akaya+Kanadaka",
+      "x": 21.5372257232666,
+      "y": -4.826983451843262
+    },
+    {
+      "name": "Average Sans",
+      "id": "average_sans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Average+Sans",
+      "x": 23.745559692382812,
+      "y": -6.659167766571045
+    },
+    {
+      "name": "Beau Rivage",
+      "id": "beau_rivage",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Beau+Rivage",
+      "x": 15.267997741699219,
+      "y": -5.466360569000244
+    },
+    {
+      "name": "Balsamiq Sans",
+      "id": "balsamiq_sans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Balsamiq+Sans",
+      "x": 23.020471572875977,
+      "y": -5.687070846557617
+    },
+    {
+      "name": "Anek Telugu",
+      "id": "anek_telugu",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Telugu",
+      "x": 19.74215316772461,
+      "y": -5.584489345550537
+    },
+    {
+      "name": "Biryani",
+      "id": "biryani",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Biryani",
+      "x": 24.789573669433594,
+      "y": -3.697936773300171
+    },
+    {
+      "name": "Bacasimentique",
+      "id": "bacasimentique",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bacasimentique",
+      "x": 16.60334014892578,
+      "y": -7.335568428039551
+    },
+    {
+      "name": "Black Ops One",
+      "id": "black_ops_one",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Black+Ops+One",
+      "x": 22.536741256713867,
+      "y": -1.6521573066711426
+    },
+    {
+      "name": "Azeret Mono",
+      "id": "azeret_mono",
+      "family": "monospace",
+      "google_fonts_url": "https://fonts.google.com/specimen/Azeret+Mono",
+      "x": 21.71756362915039,
+      "y": -3.8797690868377686
+    },
+    {
+      "name": "Borel",
+      "id": "borel",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Borel",
+      "x": 23.82610511779785,
+      "y": -4.606581687927246
+    },
+    {
+      "name": "Cairo Play",
+      "id": "cairo_play",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cairo+Play",
+      "x": 23.181394577026367,
+      "y": -3.919156551361084
+    },
+    {
+      "name": "Bona Nova",
+      "id": "bona_nova",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bona+Nova",
+      "x": 18.122859954833984,
+      "y": -9.746938705444336
+    },
+    {
+      "name": "Abeezee",
+      "id": "abeezee",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Abeezee",
+      "x": 24.58114242553711,
+      "y": -4.19850492477417
+    },
+    {
+      "name": "Abel",
+      "id": "abel",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Abel",
+      "x": 17.75788116455078,
+      "y": -4.981087684631348
+    },
+    {
+      "name": "Amaranth",
+      "id": "amaranth",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Amaranth",
+      "x": 21.90265655517578,
+      "y": -4.487709999084473
+    },
+    {
+      "name": "Caladea",
+      "id": "caladea",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Caladea",
+      "x": 17.88355827331543,
+      "y": -8.630258560180664
+    },
+    {
+      "name": "Allerta Stencil",
+      "id": "allerta_stencil",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Allerta+Stencil",
+      "x": 21.460865020751953,
+      "y": -3.8061254024505615
+    },
+    {
+      "name": "Baloo Bhaina 2",
+      "id": "baloo_bhaina_2",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baloo+Bhaina+2",
+      "x": 22.05411720275879,
+      "y": -7.818444728851318
+    },
+    {
+      "name": "Arima",
+      "id": "arima",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arima",
+      "x": 21.657363891601562,
+      "y": -6.8770222663879395
+    },
+    {
+      "name": "Beth Ellen",
+      "id": "beth_ellen",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Beth+Ellen",
+      "x": 19.031566619873047,
+      "y": -2.441924571990967
+    },
+    {
+      "name": "Assistant",
+      "id": "assistant",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Assistant",
+      "x": 18.766986846923828,
+      "y": -6.921422004699707
+    },
+    {
+      "name": "Bitcount Grid Double Ink",
+      "id": "bitcount_grid_double_ink",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Grid+Double+Ink",
+      "x": 17.511028289794922,
+      "y": 0.4614824950695038
+    },
+    {
+      "name": "Badeen Display",
+      "id": "badeen_display",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Badeen+Display",
+      "x": 20.651723861694336,
+      "y": -0.24613450467586517
+    },
+    {
+      "name": "Asimovian",
+      "id": "asimovian",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Asimovian",
+      "x": 19.239673614501953,
+      "y": -0.40000540018081665
+    },
+    {
+      "name": "Baloo Da 2",
+      "id": "baloo_da_2",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baloo+Da+2",
+      "x": 23.011268615722656,
+      "y": -7.8300604820251465
+    },
+    {
+      "name": "B612 Mono",
+      "id": "b612_mono",
+      "family": "monospace",
+      "google_fonts_url": "https://fonts.google.com/specimen/B612+Mono",
+      "x": 23.035058975219727,
+      "y": -6.1090087890625
+    },
+    {
+      "name": "Aldrich",
+      "id": "aldrich",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Aldrich",
+      "x": 23.032630920410156,
+      "y": -3.8021459579467773
+    },
+    {
+      "name": "Aref Ruqaa Ink",
+      "id": "aref_ruqaa_ink",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Aref+Ruqaa+Ink",
+      "x": 19.51122283935547,
+      "y": -8.204886436462402
+    },
+    {
+      "name": "Arsenal Sc",
+      "id": "arsenal_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arsenal+Sc",
+      "x": 18.862932205200195,
+      "y": -7.327026844024658
+    },
+    {
+      "name": "Arvo",
+      "id": "arvo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arvo",
+      "x": 25.024641036987305,
+      "y": -4.862721920013428
+    },
+    {
+      "name": "Aguafina Script",
+      "id": "aguafina_script",
+      "family": "handwriting",
+      "google_fonts_url": "https://fonts.google.com/specimen/Aguafina+Script",
+      "x": 16.580936431884766,
+      "y": -5.96278190612793
+    },
+    {
+      "name": "Bowlby One Sc",
+      "id": "bowlby_one_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bowlby+One+Sc",
+      "x": 20.67373275756836,
+      "y": -1.1447933912277222
+    },
+    {
+      "name": "Adlam Display",
+      "id": "adlam_display",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Adlam+Display",
+      "x": 22.354612350463867,
+      "y": -2.03210186958313
+    },
+    {
+      "name": "Caprasimo",
+      "id": "caprasimo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Caprasimo",
+      "x": 21.790800094604492,
+      "y": -0.797644317150116
+    },
+    {
+      "name": "Anek Kannada",
+      "id": "anek_kannada",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Kannada",
+      "x": 19.931285858154297,
+      "y": -6.161142349243164
+    },
+    {
+      "name": "Braah One",
+      "id": "braah_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Braah+One",
+      "x": 21.156309127807617,
+      "y": -2.703040599822998
+    },
+    {
+      "name": "Boldonse",
+      "id": "boldonse",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Boldonse",
+      "x": 20.186166763305664,
+      "y": -1.7771055698394775
+    },
+    {
+      "name": "Bitcount Grid Single",
+      "id": "bitcount_grid_single",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Grid+Single",
+      "x": 17.853626251220703,
+      "y": -0.31046342849731445
+    },
+    {
+      "name": "Beiruti",
+      "id": "beiruti",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Beiruti",
+      "x": 18.61107063293457,
+      "y": -6.329809188842773
+    },
+    {
+      "name": "Caveat Brush",
+      "id": "caveat_brush",
+      "family": "handwriting",
+      "google_fonts_url": "https://fonts.google.com/specimen/Caveat+Brush",
+      "x": 19.236412048339844,
+      "y": -3.9373483657836914
+    },
+    {
+      "name": "Alumni Sans Collegiate One",
+      "id": "alumni_sans_collegiate_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alumni+Sans+Collegiate+One",
+      "x": 18.921306610107422,
+      "y": -3.579806089401245
+    },
+    {
+      "name": "Chakra Petch",
+      "id": "chakra_petch",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Chakra+Petch",
+      "x": 21.94448471069336,
+      "y": -6.1062541007995605
+    },
+    {
+      "name": "Barlow Semi Condensed",
+      "id": "barlow_semi_condensed",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Barlow+Semi+Condensed",
+      "x": 18.30834197998047,
+      "y": -4.696107387542725
+    },
+    {
+      "name": "Carme",
+      "id": "carme",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Carme",
+      "x": 24.873920440673828,
+      "y": -4.715306758880615
+    },
+    {
+      "name": "Arizonia",
+      "id": "arizonia",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arizonia",
+      "x": 15.447344779968262,
+      "y": -6.747471332550049
+    },
+    {
+      "name": "Aubrey",
+      "id": "aubrey",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Aubrey",
+      "x": 18.347932815551758,
+      "y": -5.089459419250488
+    },
+    {
+      "name": "Bungee Shade",
+      "id": "bungee_shade",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bungee+Shade",
+      "x": 17.20719337463379,
+      "y": -4.653548717498779
+    },
+    {
+      "name": "Allan",
+      "id": "allan",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Allan",
+      "x": 16.424718856811523,
+      "y": -7.599412441253662
+    },
+    {
+      "name": "Almarai",
+      "id": "almarai",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Almarai",
+      "x": 24.12346076965332,
+      "y": -5.633124351501465
+    },
+    {
+      "name": "Cabin",
+      "id": "cabin",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cabin",
+      "x": 22.44768714904785,
+      "y": -5.188098907470703
+    },
+    {
+      "name": "Castoro",
+      "id": "castoro",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Castoro",
+      "x": 20.40385627746582,
+      "y": -10.087404251098633
+    },
+    {
+      "name": "Belanosima",
+      "id": "belanosima",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Belanosima",
+      "x": 20.7187557220459,
+      "y": -4.848654747009277
+    },
+    {
+      "name": "Bricolage Grotesque",
+      "id": "bricolage_grotesque",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bricolage+Grotesque",
+      "x": 22.61180305480957,
+      "y": -4.70428991317749
+    },
+    {
+      "name": "Are You Serious",
+      "id": "are_you_serious",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Are+You+Serious",
+      "x": 15.627021789550781,
+      "y": -5.892740249633789
+    },
+    {
+      "name": "Afacad",
+      "id": "afacad",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Afacad",
+      "x": 17.835962295532227,
+      "y": -7.096208095550537
+    },
+    {
+      "name": "Anek Gujarati",
+      "id": "anek_gujarati",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Gujarati",
+      "x": 18.848543167114258,
+      "y": -5.986158847808838
+    },
+    {
+      "name": "Aclonica",
+      "id": "aclonica",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Aclonica",
+      "x": 19.861370086669922,
+      "y": -0.49223336577415466
+    },
+    {
+      "name": "Andada Pro",
+      "id": "andada_pro",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Andada+Pro",
+      "x": 19.1613712310791,
+      "y": -10.264761924743652
+    },
+    {
+      "name": "B612",
+      "id": "b612",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/B612",
+      "x": 21.332172393798828,
+      "y": -5.605536460876465
+    },
+    {
+      "name": "Albert Sans",
+      "id": "albert_sans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Albert+Sans",
+      "x": 23.795520782470703,
+      "y": -3.814154624938965
+    },
+    {
+      "name": "Bubbler One",
+      "id": "bubbler_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bubbler+One",
+      "x": 16.138294219970703,
+      "y": -7.374259948730469
+    },
+    {
+      "name": "Alfa Slab One",
+      "id": "alfa_slab_one",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alfa+Slab+One",
+      "x": 21.512500762939453,
+      "y": -0.6457157731056213
+    },
+    {
+      "name": "Baskervville",
+      "id": "baskervville",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baskervville",
+      "x": 21.01013946533203,
+      "y": -9.273618698120117
+    },
+    {
+      "name": "Bebas Neue",
+      "id": "bebas_neue",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bebas+Neue",
+      "x": 17.868955612182617,
+      "y": -3.7222964763641357
+    },
+    {
+      "name": "Benne",
+      "id": "benne",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Benne",
+      "x": 17.892671585083008,
+      "y": -9.818638801574707
+    },
+    {
+      "name": "Alike",
+      "id": "alike",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alike",
+      "x": 20.023128509521484,
+      "y": -10.421673774719238
+    },
+    {
+      "name": "Cardo",
+      "id": "cardo",
+      "family": "serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cardo",
+      "x": 20.306049346923828,
+      "y": -8.33852767944336
+    },
+    {
+      "name": "Architects Daughter",
+      "id": "architects_daughter",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Architects+Daughter",
+      "x": 21.24519920349121,
+      "y": -8.08905029296875
+    },
+    {
+      "name": "Cabin Sketch",
+      "id": "cabin_sketch",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cabin+Sketch",
+      "x": 17.003170013427734,
+      "y": -6.8975982666015625
+    },
+    {
+      "name": "Alef",
+      "id": "alef",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alef",
+      "x": 24.483924865722656,
+      "y": -5.49644136428833
+    },
+    {
+      "name": "Bangers",
+      "id": "bangers",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bangers",
+      "x": 19.234437942504883,
+      "y": -3.465296745300293
+    },
+    {
+      "name": "Baloo Thambi 2",
+      "id": "baloo_thambi_2",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baloo+Thambi+2",
+      "x": 21.80781364440918,
+      "y": -7.875688552856445
+    },
+    {
+      "name": "Calligraffitti",
+      "id": "calligraffitti",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Calligraffitti",
+      "x": 16.124544143676758,
+      "y": -6.571818828582764
+    },
+    {
+      "name": "Alex Brush",
+      "id": "alex_brush",
+      "family": "handwriting",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alex+Brush",
+      "x": 14.987281799316406,
+      "y": -5.5780863761901855
+    },
+    {
+      "name": "Asta Sans",
+      "id": "asta_sans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Asta+Sans",
+      "x": 22.94811248779297,
+      "y": -5.976537704467773
+    },
+    {
+      "name": "Basic",
+      "id": "basic",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Basic",
+      "x": 21.044729232788086,
+      "y": -4.437438488006592
+    },
+    {
+      "name": "Atkinson Hyperlegible Next",
+      "id": "atkinson_hyperlegible_next",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Atkinson+Hyperlegible+Next",
+      "x": 22.040233612060547,
+      "y": -4.846309661865234
+    },
+    {
+      "name": "Angkor",
+      "id": "angkor",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Angkor",
+      "x": 20.85524559020996,
+      "y": -0.014288547448813915
+    },
+    {
+      "name": "Boogaloo",
+      "id": "boogaloo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Boogaloo",
+      "x": 19.820871353149414,
+      "y": -3.13828706741333
+    },
+    {
+      "name": "Alumni Sans Sc",
+      "id": "alumni_sans_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alumni+Sans+Sc",
+      "x": 16.54025650024414,
+      "y": -4.281049728393555
+    },
+    {
+      "name": "Averia Sans Libre",
+      "id": "averia_sans_libre",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Averia+Sans+Libre",
+      "x": 21.177377700805664,
+      "y": -5.337528228759766
+    },
+    {
+      "name": "Biz Udgothic",
+      "id": "biz_udgothic",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Biz+Udgothic",
+      "x": 18.47075653076172,
+      "y": -3.9350242614746094
+    },
+    {
+      "name": "Cactus Classical Serif",
+      "id": "cactus_classical_serif",
+      "family": "serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cactus+Classical+Serif",
+      "x": 18.429519653320312,
+      "y": -9.250237464904785
+    },
+    {
+      "name": "Baskervville Sc",
+      "id": "baskervville_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baskervville+Sc",
+      "x": 21.187152862548828,
+      "y": -9.828973770141602
+    },
+    {
+      "name": "Biz Udpgothic",
+      "id": "biz_udpgothic",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Biz+Udpgothic",
+      "x": 24.917390823364258,
+      "y": -3.4352715015411377
+    },
+    {
+      "name": "Bytesized",
+      "id": "bytesized",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bytesized",
+      "x": 19.65236473083496,
+      "y": -4.436420917510986
+    },
+    {
+      "name": "Cambo",
+      "id": "cambo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cambo",
+      "x": 19.6998348236084,
+      "y": -9.475263595581055
+    },
+    {
+      "name": "Almendra Display",
+      "id": "almendra_display",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Almendra+Display",
+      "x": 16.643239974975586,
+      "y": -6.247558116912842
+    },
+    {
+      "name": "Average",
+      "id": "average",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Average",
+      "x": 19.847530364990234,
+      "y": -9.892707824707031
+    },
+    {
+      "name": "Carrois Gothic",
+      "id": "carrois_gothic",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Carrois+Gothic",
+      "x": 19.901945114135742,
+      "y": -4.897567272186279
+    },
+    {
+      "name": "Biorhyme",
+      "id": "biorhyme",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Biorhyme",
+      "x": 24.187475204467773,
+      "y": -3.3428378105163574
+    },
+    {
+      "name": "Berkshire Swash",
+      "id": "berkshire_swash",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Berkshire+Swash",
+      "x": 21.21091651916504,
+      "y": -3.136693000793457
+    },
+    {
+      "name": "Bad Script",
+      "id": "bad_script",
+      "family": "handwriting",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bad+Script",
+      "x": 16.321144104003906,
+      "y": -5.153273582458496
+    },
+    {
+      "name": "Bitcount Prop Double Ink",
+      "id": "bitcount_prop_double_ink",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Prop+Double+Ink",
+      "x": 17.084135055541992,
+      "y": 0.1654723882675171
+    },
+    {
+      "name": "Agbalumo",
+      "id": "agbalumo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Agbalumo",
+      "x": 21.516958236694336,
+      "y": -1.943345546722412
+    },
+    {
+      "name": "Anta",
+      "id": "anta",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anta",
+      "x": 23.323423385620117,
+      "y": -5.006962299346924
+    },
+    {
+      "name": "Artifika",
+      "id": "artifika",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Artifika",
+      "x": 19.625858306884766,
+      "y": -8.712949752807617
+    },
+    {
+      "name": "Bai Jamjuree",
+      "id": "bai_jamjuree",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bai+Jamjuree",
+      "x": 22.658227920532227,
+      "y": -6.281856060028076
+    },
+    {
+      "name": "Bitcount Single Ink",
+      "id": "bitcount_single_ink",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Single+Ink",
+      "x": 17.931718826293945,
+      "y": -0.6211944818496704
+    },
+    {
+      "name": "Baloo 2",
+      "id": "baloo_2",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baloo+2",
+      "x": 22.44927406311035,
+      "y": -7.514186859130859
+    },
+    {
+      "name": "Annie Use Your Telescope",
+      "id": "annie_use_your_telescope",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Annie+Use+Your+Telescope",
+      "x": 17.199050903320312,
+      "y": -6.687832832336426
+    },
+    {
+      "name": "Agdasima",
+      "id": "agdasima",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Agdasima",
+      "x": 17.61175537109375,
+      "y": -4.068910121917725
+    },
+    {
+      "name": "Atkinson Hyperlegible",
+      "id": "atkinson_hyperlegible",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Atkinson+Hyperlegible",
+      "x": 22.29834747314453,
+      "y": -4.985901832580566
+    },
+    {
+      "name": "Allerta",
+      "id": "allerta",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Allerta",
+      "x": 21.451335906982422,
+      "y": -3.6136300563812256
+    },
+    {
+      "name": "Asul",
+      "id": "asul",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Asul",
+      "x": 20.711162567138672,
+      "y": -8.710515022277832
+    },
+    {
+      "name": "Baloo Bhaijaan 2",
+      "id": "baloo_bhaijaan_2",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baloo+Bhaijaan+2",
+      "x": 22.06007957458496,
+      "y": -7.522099018096924
+    },
+    {
+      "name": "Amita",
+      "id": "amita",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Amita",
+      "x": 17.209745407104492,
+      "y": -1.7015841007232666
+    },
+    {
+      "name": "Astloch",
+      "id": "astloch",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Astloch",
+      "x": 16.062471389770508,
+      "y": -5.617806434631348
+    },
+    {
+      "name": "Bonheur Royale",
+      "id": "bonheur_royale",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bonheur+Royale",
+      "x": 15.584623336791992,
+      "y": -7.091217041015625
+    },
+    {
+      "name": "Bahianita",
+      "id": "bahianita",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bahianita",
+      "x": 16.54690933227539,
+      "y": -3.694916009902954
+    },
+    {
+      "name": "Amarante",
+      "id": "amarante",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Amarante",
+      "x": 21.074485778808594,
+      "y": -5.431578636169434
+    },
+    {
+      "name": "Birthstone",
+      "id": "birthstone",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Birthstone",
+      "x": 15.736165046691895,
+      "y": -6.779961109161377
+    },
+    {
+      "name": "Bungee Hairline",
+      "id": "bungee_hairline",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bungee+Hairline",
+      "x": 16.096961975097656,
+      "y": -6.074085235595703
+    },
+    {
+      "name": "Anek Gurmukhi",
+      "id": "anek_gurmukhi",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Gurmukhi",
+      "x": 18.717300415039062,
+      "y": -6.102351188659668
+    },
+    {
+      "name": "Bevan",
+      "id": "bevan",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bevan",
+      "x": 21.574953079223633,
+      "y": -0.31053969264030457
+    },
+    {
+      "name": "Belleza",
+      "id": "belleza",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Belleza",
+      "x": 19.45440673828125,
+      "y": -7.8840131759643555
+    },
+    {
+      "name": "Caesar Dressing",
+      "id": "caesar_dressing",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Caesar+Dressing",
+      "x": 20.18159294128418,
+      "y": -4.050620079040527
+    },
+    {
+      "name": "Cedarville Cursive",
+      "id": "cedarville_cursive",
+      "family": "handwriting",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cedarville+Cursive",
+      "x": 16.8320255279541,
+      "y": -4.985641956329346
+    },
+    {
+      "name": "Cascadia Code",
+      "id": "cascadia_code",
+      "family": "monospace",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cascadia+Code",
+      "x": 20.441112518310547,
+      "y": -4.06734561920166
+    },
+    {
+      "name": "Bungee",
+      "id": "bungee",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bungee",
+      "x": 20.856475830078125,
+      "y": -0.5865234732627869
+    },
+    {
+      "name": "Big Shoulders",
+      "id": "big_shoulders",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Big+Shoulders",
+      "x": 17.095279693603516,
+      "y": -2.92136549949646
+    },
+    {
+      "name": "Buenard",
+      "id": "buenard",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Buenard",
+      "x": 18.869497299194336,
+      "y": -10.033001899719238
+    },
+    {
+      "name": "Alegreya Sans",
+      "id": "alegreya_sans",
+      "family": "serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alegreya+Sans",
+      "x": 20.300996780395508,
+      "y": -6.641489028930664
+    },
+    {
+      "name": "Antic Slab",
+      "id": "antic_slab",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Antic+Slab",
+      "x": 22.1838436126709,
+      "y": -6.957996845245361
+    },
+    {
+      "name": "Advent Pro",
+      "id": "advent_pro",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Advent+Pro",
+      "x": 18.141855239868164,
+      "y": -7.8316264152526855
+    },
+    {
+      "name": "Capriola",
+      "id": "capriola",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Capriola",
+      "x": 23.789642333984375,
+      "y": -2.830886125564575
+    },
+    {
+      "name": "Bitter",
+      "id": "bitter",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitter",
+      "x": 23.30760383605957,
+      "y": -6.923764705657959
+    },
+    {
+      "name": "Butcherman",
+      "id": "butcherman",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Butcherman",
+      "x": 17.857656478881836,
+      "y": -2.5151383876800537
+    },
+    {
+      "name": "Allura",
+      "id": "allura",
+      "family": "handwriting",
+      "google_fonts_url": "https://fonts.google.com/specimen/Allura",
+      "x": 14.982673645019531,
+      "y": -4.924966335296631
+    },
+    {
+      "name": "Big Shoulders Stencil",
+      "id": "big_shoulders_stencil",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Big+Shoulders+Stencil",
+      "x": 17.181045532226562,
+      "y": -3.0563745498657227
+    },
+    {
+      "name": "Alegreya Sc",
+      "id": "alegreya_sc",
+      "family": "serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alegreya+Sc",
+      "x": 17.37143898010254,
+      "y": -9.784417152404785
+    },
+    {
+      "name": "Averia Libre",
+      "id": "averia_libre",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Averia+Libre",
+      "x": 22.089916229248047,
+      "y": -6.2136921882629395
+    },
+    {
+      "name": "Bilbo",
+      "id": "bilbo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bilbo",
+      "x": 15.704448699951172,
+      "y": -7.2287278175354
+    },
+    {
+      "name": "Battambang",
+      "id": "battambang",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Battambang",
+      "x": 24.13847541809082,
+      "y": -6.134006977081299
+    },
+    {
+      "name": "Atma",
+      "id": "atma",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Atma",
+      "x": 23.992830276489258,
+      "y": -6.292703151702881
+    },
+    {
+      "name": "Antic Didone",
+      "id": "antic_didone",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Antic+Didone",
+      "x": 17.430044174194336,
+      "y": -6.857850551605225
+    },
+    {
+      "name": "Bitcount Prop Double",
+      "id": "bitcount_prop_double",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Prop+Double",
+      "x": 17.106901168823242,
+      "y": -0.6386945247650146
+    },
+    {
+      "name": "Alumni Sans",
+      "id": "alumni_sans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alumni+Sans",
+      "x": 16.732097625732422,
+      "y": -4.37276554107666
+    },
+    {
+      "name": "Brunoce",
+      "id": "brunoce",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Brunoce",
+      "x": 15.54136848449707,
+      "y": -4.220706462860107
+    },
+    {
+      "name": "Adamina",
+      "id": "adamina",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Adamina",
+      "x": 21.088266372680664,
+      "y": -7.7849650382995605
+    },
+    {
+      "name": "Baloo Tammudu 2",
+      "id": "baloo_tammudu_2",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baloo+Tammudu+2",
+      "x": 22.303668975830078,
+      "y": -8.426968574523926
+    },
+    {
+      "name": "Caramel",
+      "id": "caramel",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Caramel",
+      "x": 15.850841522216797,
+      "y": -6.1333441734313965
+    },
+    {
+      "name": "Atomicge",
+      "id": "atomicge",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Atomicge",
+      "x": 21.877777099609375,
+      "y": -3.5024640560150146
+    },
+    {
+      "name": "Ancizar Sans",
+      "id": "ancizar_sans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Ancizar+Sans",
+      "x": 19.752065658569336,
+      "y": -7.119527339935303
+    },
+    {
+      "name": "Alata",
+      "id": "alata",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alata",
+      "x": 22.926610946655273,
+      "y": -3.6643505096435547
+    },
+    {
+      "name": "Blaka",
+      "id": "blaka",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Blaka",
+      "x": 19.51890754699707,
+      "y": -3.004938840866089
+    },
+    {
+      "name": "Baloo Paaji 2",
+      "id": "baloo_paaji_2",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baloo+Paaji+2",
+      "x": 22.315481185913086,
+      "y": -8.310994148254395
+    },
+    {
+      "name": "Aboreto",
+      "id": "aboreto",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Aboreto",
+      "x": 20.536846160888672,
+      "y": -7.52553129196167
+    },
+    {
+      "name": "Carlito",
+      "id": "carlito",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Carlito",
+      "x": 21.281217575073242,
+      "y": -7.098779201507568
+    },
+    {
+      "name": "Bitcount Grid Single Ink",
+      "id": "bitcount_grid_single_ink",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Grid+Single+Ink",
+      "x": 17.839941024780273,
+      "y": -0.008455192670226097
+    },
+    {
+      "name": "Bahiana",
+      "id": "bahiana",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bahiana",
+      "x": 16.575590133666992,
+      "y": -3.7378668785095215
+    },
+    {
+      "name": "Amiko",
+      "id": "amiko",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Amiko",
+      "x": 24.26144027709961,
+      "y": -4.247131824493408
+    },
+    {
+      "name": "Anybody",
+      "id": "anybody",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anybody",
+      "x": 23.013492584228516,
+      "y": -4.421940803527832
+    },
+    {
+      "name": "Carrois Gothic Sc",
+      "id": "carrois_gothic_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Carrois+Gothic+Sc",
+      "x": 19.822664260864258,
+      "y": -4.619467258453369
+    },
+    {
+      "name": "Alatsi",
+      "id": "alatsi",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alatsi",
+      "x": 19.052143096923828,
+      "y": -4.417135238647461
+    },
+    {
+      "name": "Atkinson Hyperlegible Mono",
+      "id": "atkinson_hyperlegible_mono",
+      "family": "monospace",
+      "google_fonts_url": "https://fonts.google.com/specimen/Atkinson+Hyperlegible+Mono",
+      "x": 21.013835906982422,
+      "y": -5.903766632080078
+    },
+    {
+      "name": "Barrio",
+      "id": "barrio",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Barrio",
+      "x": 16.895658493041992,
+      "y": -8.007171630859375
+    },
+    {
+      "name": "Caveat",
+      "id": "caveat",
+      "family": "handwriting",
+      "google_fonts_url": "https://fonts.google.com/specimen/Caveat",
+      "x": 15.944704055786133,
+      "y": -7.713915824890137
+    },
+    {
+      "name": "Bubblegum Sans",
+      "id": "bubblegum_sans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bubblegum+Sans",
+      "x": 20.563289642333984,
+      "y": -4.343777656555176
+    },
+    {
+      "name": "Allison",
+      "id": "allison",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Allison",
+      "x": 15.022273063659668,
+      "y": -6.174474239349365
+    },
+    {
+      "name": "Anton Sc",
+      "id": "anton_sc",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anton+Sc",
+      "x": 18.491214752197266,
+      "y": -2.559946298599243
+    },
+    {
+      "name": "Alexandria",
+      "id": "alexandria",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alexandria",
+      "x": 23.743505477905273,
+      "y": -2.6539440155029297
+    },
+    {
+      "name": "Ceviche One",
+      "id": "ceviche_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Ceviche+One",
+      "x": 20.516429901123047,
+      "y": -3.8042242527008057
+    },
+    {
+      "name": "Arsenal",
+      "id": "arsenal",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arsenal",
+      "x": 18.58235740661621,
+      "y": -7.172868728637695
+    },
+    {
+      "name": "Bayon",
+      "id": "bayon",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bayon",
+      "x": 18.013530731201172,
+      "y": -3.4357454776763916
+    },
+    {
+      "name": "Alkalami",
+      "id": "alkalami",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alkalami",
+      "x": 18.847293853759766,
+      "y": -9.315666198730469
+    },
+    {
+      "name": "Biz Udpmincho",
+      "id": "biz_udpmincho",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Biz+Udpmincho",
+      "x": 25.27707290649414,
+      "y": -5.603499889373779
+    },
+    {
+      "name": "Bodoni Moda",
+      "id": "bodoni_moda",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bodoni+Moda",
+      "x": 18.617900848388672,
+      "y": -8.028144836425781
+    },
+    {
+      "name": "Besley",
+      "id": "besley",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Besley",
+      "x": 24.415678024291992,
+      "y": -5.008090019226074
+    },
+    {
+      "name": "Barriecito",
+      "id": "barriecito",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Barriecito",
+      "x": 17.228063583374023,
+      "y": -7.839105606079102
+    },
+    {
+      "name": "Bitcount Single",
+      "id": "bitcount_single",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Single",
+      "x": 17.378244400024414,
+      "y": -0.7432816624641418
+    },
+    {
+      "name": "Alumni Sans Pinstripe",
+      "id": "alumni_sans_pinstripe",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alumni+Sans+Pinstripe",
+      "x": 15.780563354492188,
+      "y": -5.802762031555176
+    },
+    {
+      "name": "Biz Udmincho",
+      "id": "biz_udmincho",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Biz+Udmincho",
+      "x": 16.8121337890625,
+      "y": -3.3362252712249756
+    },
+    {
+      "name": "Benchnine",
+      "id": "benchnine",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Benchnine",
+      "x": 17.077804565429688,
+      "y": -3.8785977363586426
+    },
+    {
+      "name": "Anek Tamil",
+      "id": "anek_tamil",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Tamil",
+      "x": 19.905885696411133,
+      "y": -5.835819244384766
+    },
+    {
+      "name": "Blinker",
+      "id": "blinker",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Blinker",
+      "x": 21.772686004638672,
+      "y": -6.7732696533203125
+    },
+    {
+      "name": "Bentham",
+      "id": "bentham",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bentham",
+      "x": 18.146596908569336,
+      "y": -8.387253761291504
+    },
+    {
+      "name": "Autour One",
+      "id": "autour_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Autour+One",
+      "x": 25.084901809692383,
+      "y": -3.8402962684631348
+    },
+    {
+      "name": "Cabin Condensed",
+      "id": "cabin_condensed",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cabin+Condensed",
+      "x": 19.268564224243164,
+      "y": -4.791539669036865
+    },
+    {
+      "name": "Bilbo Swash Caps",
+      "id": "bilbo_swash_caps",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bilbo+Swash+Caps",
+      "x": 16.41958236694336,
+      "y": -7.31458044052124
+    },
+    {
+      "name": "Aref Ruqaa",
+      "id": "aref_ruqaa",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Aref+Ruqaa",
+      "x": 19.727100372314453,
+      "y": -8.110301971435547
+    },
+    {
+      "name": "Cairo",
+      "id": "cairo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cairo",
+      "x": 22.79924774169922,
+      "y": -4.124084949493408
+    },
+    {
+      "name": "Alkatra",
+      "id": "alkatra",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alkatra",
+      "x": 22.097200393676758,
+      "y": -4.962743759155273
+    },
+    {
+      "name": "Barlow",
+      "id": "barlow",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Barlow",
+      "x": 20.214824676513672,
+      "y": -5.760263919830322
+    },
+    {
+      "name": "Cagliostro",
+      "id": "cagliostro",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cagliostro",
+      "x": 24.69200897216797,
+      "y": -3.1803441047668457
+    },
+    {
+      "name": "Bitcount Ink",
+      "id": "bitcount_ink",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Ink",
+      "x": 17.3082332611084,
+      "y": 0.40409553050994873
+    },
+    {
+      "name": "Bungee Tint",
+      "id": "bungee_tint",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bungee+Tint",
+      "x": 21.28632164001465,
+      "y": -1.318466305732727
+    },
+    {
+      "name": "Bakbak One",
+      "id": "bakbak_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bakbak+One",
+      "x": 21.92405891418457,
+      "y": -2.854964256286621
+    },
+    {
+      "name": "Anaheim",
+      "id": "anaheim",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anaheim",
+      "x": 18.546825408935547,
+      "y": -5.666948318481445
+    },
+    {
+      "name": "Akshar",
+      "id": "akshar",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Akshar",
+      "x": 18.005294799804688,
+      "y": -3.8710296154022217
+    },
+    {
+      "name": "Cascadia Mono",
+      "id": "cascadia_mono",
+      "family": "monospace",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cascadia+Mono",
+      "x": 20.30777359008789,
+      "y": -3.9027891159057617
+    },
+    {
+      "name": "Akronim",
+      "id": "akronim",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Akronim",
+      "x": 17.307586669921875,
+      "y": -6.330279350280762
+    },
+    {
+      "name": "Agu Display",
+      "id": "agu_display",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Agu+Display",
+      "x": 19.590599060058594,
+      "y": -6.791477203369141
+    },
+    {
+      "name": "Antonio",
+      "id": "antonio",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Antonio",
+      "x": 17.53076171875,
+      "y": -3.3967697620391846
+    },
+    {
+      "name": "Arimo",
+      "id": "arimo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arimo",
+      "x": 24.819883346557617,
+      "y": -4.613128662109375
+    },
+    {
+      "name": "Bagel Fat One",
+      "id": "bagel_fat_one",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bagel+Fat+One",
+      "x": 21.558013916015625,
+      "y": -1.218587040901184
+    },
+    {
+      "name": "Bellota Text",
+      "id": "bellota_text",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bellota+Text",
+      "x": 21.47699737548828,
+      "y": -8.807967185974121
+    },
+    {
+      "name": "Candal",
+      "id": "candal",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Candal",
+      "x": 22.059022903442383,
+      "y": -1.7491744756698608
+    },
+    {
+      "name": "Anek Bangla",
+      "id": "anek_bangla",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Bangla",
+      "x": 19.23285675048828,
+      "y": -5.678542613983154
+    },
+    {
+      "name": "Babylonica",
+      "id": "babylonica",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Babylonica",
+      "x": 15.224921226501465,
+      "y": -4.503042697906494
+    },
+    {
+      "name": "Blaka Ink",
+      "id": "blaka_ink",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Blaka+Ink",
+      "x": 19.087324142456055,
+      "y": -2.785191059112549
+    },
+    {
+      "name": "Athiti",
+      "id": "athiti",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Athiti",
+      "x": 19.118690490722656,
+      "y": -6.860150337219238
+    },
+    {
+      "name": "Antic",
+      "id": "antic",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Antic",
+      "x": 21.03575325012207,
+      "y": -6.856987476348877
+    },
+    {
+      "name": "Averia Gruesa Libre",
+      "id": "averia_gruesa_libre",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Averia+Gruesa+Libre",
+      "x": 20.79208755493164,
+      "y": -5.69139289855957
+    },
+    {
+      "name": "Baloo Bhai 2",
+      "id": "baloo_bhai_2",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Baloo+Bhai+2",
+      "x": 22.186534881591797,
+      "y": -8.231916427612305
+    },
+    {
+      "name": "Asap",
+      "id": "asap",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Asap",
+      "x": 22.905757904052734,
+      "y": -5.3783040046691895
+    },
+    {
+      "name": "Black Han Sans",
+      "id": "black_han_sans",
+      "family": "display",
+      "google_fonts_url": "https://fonts.google.com/specimen/Black+Han+Sans",
+      "x": 21.315765380859375,
+      "y": -0.47961828112602234
+    },
+    {
+      "name": "Bodoni Moda Sc",
+      "id": "bodoni_moda_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bodoni+Moda+Sc",
+      "x": 18.4742431640625,
+      "y": -8.070531845092773
+    },
+    {
+      "name": "Abhaya Libre",
+      "id": "abhaya_libre",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Abhaya+Libre",
+      "x": 16.599609375,
+      "y": -8.718674659729004
+    },
+    {
+      "name": "Bokor",
+      "id": "bokor",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bokor",
+      "x": 18.343984603881836,
+      "y": -2.9644086360931396
+    },
+    {
+      "name": "Cal Sans",
+      "id": "cal_sans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Cal+Sans",
+      "x": 22.68681526184082,
+      "y": -2.310499429702759
+    },
+    {
+      "name": "Bona Nova Sc",
+      "id": "bona_nova_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bona+Nova+Sc",
+      "x": 18.507049560546875,
+      "y": -9.78375244140625
+    },
+    {
+      "name": "Bellota",
+      "id": "bellota",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bellota",
+      "x": 17.410242080688477,
+      "y": -6.100600719451904
+    },
+    {
+      "name": "Acme",
+      "id": "acme",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Acme",
+      "x": 22.13041877746582,
+      "y": -3.7585763931274414
+    },
+    {
+      "name": "Amatic Sc",
+      "id": "amatic_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Amatic+Sc",
+      "x": 16.485231399536133,
+      "y": -5.398030757904053
+    },
+    {
+      "name": "Brawler",
+      "id": "brawler",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Brawler",
+      "x": 21.43292808532715,
+      "y": -9.185400009155273
+    },
+    {
+      "name": "Birthstone Bounce",
+      "id": "birthstone_bounce",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Birthstone+Bounce",
+      "x": 15.024459838867188,
+      "y": -6.570058822631836
+    },
+    {
+      "name": "Alikengular",
+      "id": "alikengular",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Alikengular",
+      "x": 19.844722747802734,
+      "y": -10.29491138458252
+    },
+    {
+      "name": "Arbutus Slab",
+      "id": "arbutus_slab",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arbutus+Slab",
+      "x": 24.78438949584961,
+      "y": -6.27247953414917
+    },
+    {
+      "name": "Bungee Spice",
+      "id": "bungee_spice",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bungee+Spice",
+      "x": 20.85809326171875,
+      "y": -1.1252802610397339
+    },
+    {
+      "name": "Annapurna Sil",
+      "id": "annapurna_sil",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Annapurna+Sil",
+      "x": 17.107789993286133,
+      "y": -9.338480949401855
+    },
+    {
+      "name": "Ar One Sans",
+      "id": "ar_one_sans",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Ar+One+Sans",
+      "x": 23.89826202392578,
+      "y": -5.390315055847168
+    },
+    {
+      "name": "Brygada 1918",
+      "id": "brygada_1918",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Brygada+1918",
+      "x": 19.928668975830078,
+      "y": -9.552244186401367
+    },
+    {
+      "name": "Asar",
+      "id": "asar",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Asar",
+      "x": 19.002246856689453,
+      "y": -8.500955581665039
+    },
+    {
+      "name": "Arapey",
+      "id": "arapey",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Arapey",
+      "x": 17.51120948791504,
+      "y": -8.063387870788574
+    },
+    {
+      "name": "Brunoce Sc",
+      "id": "brunoce_sc",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Brunoce+Sc",
+      "x": 15.582780838012695,
+      "y": -4.493573188781738
+    },
+    {
+      "name": "Bungee Outline",
+      "id": "bungee_outline",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bungee+Outline",
+      "x": 16.014991760253906,
+      "y": -5.308678150177002
+    },
+    {
+      "name": "Ballet",
+      "id": "ballet",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Ballet",
+      "x": 15.107850074768066,
+      "y": -5.343606948852539
+    },
+    {
+      "name": "Belgrano",
+      "id": "belgrano",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Belgrano",
+      "x": 25.045072555541992,
+      "y": -5.672089576721191
+    },
+    {
+      "name": "Anonymous Pro",
+      "id": "anonymous_pro",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anonymous+Pro",
+      "x": 18.98392105102539,
+      "y": -6.470157623291016
+    },
+    {
+      "name": "Bitcount Prop Single Ink",
+      "id": "bitcount_prop_single_ink",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Prop+Single+Ink",
+      "x": 18.23072052001953,
+      "y": -0.23990213871002197
+    },
+    {
+      "name": "Bigelow Rules",
+      "id": "bigelow_rules",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bigelow+Rules",
+      "x": 16.611814498901367,
+      "y": -4.771378517150879
+    },
+    {
+      "name": "Almendra",
+      "id": "almendra",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Almendra",
+      "x": 16.88104820251465,
+      "y": -8.615220069885254
+    },
+    {
+      "name": "Bitcount Grid Double",
+      "id": "bitcount_grid_double",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bitcount+Grid+Double",
+      "x": 16.82975196838379,
+      "y": -0.046033602207899094
+    },
+    {
+      "name": "Aleo",
+      "id": "aleo",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Aleo",
+      "x": 25.546146392822266,
+      "y": -4.146620273590088
+    },
+    {
+      "name": "Armata",
+      "id": "armata",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Armata",
+      "x": 23.77332878112793,
+      "y": -5.257111072540283
+    },
+    {
+      "name": "Anek Devanagari",
+      "id": "anek_devanagari",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Anek+Devanagari",
+      "x": 19.55430030822754,
+      "y": -5.812459468841553
+    },
+    {
+      "name": "Asap Condensed",
+      "id": "asap_condensed",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Asap+Condensed",
+      "x": 18.80503273010254,
+      "y": -4.8164801597595215
+    },
+    {
+      "name": "Bellefair",
+      "id": "bellefair",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Bellefair",
+      "x": 19.255563735961914,
+      "y": -9.6854887008667
+    },
+    {
+      "name": "Archivo Narrow",
+      "id": "archivo_narrow",
+      "family": "sans-serif",
+      "google_fonts_url": "https://fonts.google.com/specimen/Archivo+Narrow",
+      "x": 20.823579788208008,
+      "y": -5.006474018096924
+    }
+  ]
+}

data/vision.csv

@@ -0,0 +1,186 @@
+subset_name,total_images,total_samples,total_turns,question_total_tokens,answer_total_tokens,eagle_cathegory
+coco_colors,118287,118287,118287,1301157,6376672,Captioning & Knowledge
+densefusion_1m,1058751,1058751,1058751,10692478,263718217,Captioning & Knowledge
+face_emotion,797,797,797,8767,8066,Captioning & Knowledge
+google-landmarks,299993,299993,842127,6194978,10202980,Captioning & Knowledge
+image_textualization(filtered),99573,99573,99573,917577,19374090,Captioning & Knowledge
+laion_gpt4v,9301,9301,9301,93950,1875283,Captioning & Knowledge
+localized_narratives,199998,199998,199998,2167179,8021473,Captioning & Knowledge
+sharegpt4o,57284,57284,57284,558647,36555323,Captioning & Knowledge
+sharegpt4v(coco),50017,50017,50017,460893,9825387,Captioning & Knowledge
+sharegpt4v(knowledge),1988,1988,1988,18250,293850,Captioning & Knowledge
+sharegpt4v(llava),29986,29986,29986,275783,6175899,Captioning & Knowledge
+sharegpt4v(sam),8990,8990,8990,82874,1668797,Captioning & Knowledge
+textcaps,21906,21906,21906,240966,355991,Captioning & Knowledge
+chart2text,26961,26961,30215,342215,2670580,Chart & Table
+chartqa,18265,18265,28287,625569,134793,Chart & Table
+CoSyn_400k_chart,116814,116814,1085882,17617591,57641030,Chart & Table
+CoSyn_400k_table,46518,46518,416519,6280455,23335054,Chart & Table
+dvqa,200000,200000,2325316,44603372,5477966,Chart & Table
+figureqa,100000,100000,1327368,18515153,2654736,Chart & Table
+figureqa(mathv360k),17587,17587,17587,722959,97404,Chart & Table
+finqa,5276,5276,6251,5552943,224015,Chart & Table
+hitab,2500,2500,7782,177999,335013,Chart & Table
+lrv_chart,1776,1776,5372,76477,158711,Chart & Table
+mmc_instruct,168178,168178,168178,50008824,74581055,Chart & Table
+multihiertt,30875,7619,7830,218840,244744,Chart & Table
+plotqa,157070,157070,20249479,738371054,118122387,Chart & Table
+robut_sqa,8514,8514,34141,368957,1794570,Chart & Table
+robut_wikisql,74989,74989,86202,1454920,9276100,Chart & Table
+robut_wtq,38246,38246,44096,587040,6415830,Chart & Table
+SynthChartNet,500000,500000,500000,2169240,67392223,Chart & Table
+tabmwp,22722,22722,23021,639639,1883243,Chart & Table
+tabmwp(mathv360k),22452,22452,22452,963498,158042,Chart & Table
+tat_dqa,2448,2207,13251,320356,1177852,Chart & Table
+tat_qa,2199,2199,13215,989419,254790,Chart & Table
+Unichart,611925,611925,6898324,96702288,211989247,Chart & Table
+vistext,9969,9969,9969,88770,1191127,Chart & Table
+vqaonbd,39986,39986,1254165,36066807,5620523,Chart & Table
+alfworldgpt,45073,45073,45073,17864033,6276573,General VQA
+allava_laion,468664,468664,937328,18654303,145799426,General VQA
+allava_vflan,177078,177078,387872,12444711,55305642,General VQA
+cambrian(filtered)_processed,83123,83124,98534,1410321,5503211,General VQA
+chinesememe,54212,54212,54212,538938,21122723,General VQA
+cocoqa,46287,46287,78736,1136238,212480,General VQA
+CoSyn_400k_graphic,26968,26968,26968,1678862,8235679,General VQA
+datik,220537,222385,222385,2234054,187757952,General VQA
+datikz,47441,47974,48296,441040,59116193,General VQA
+drivelm,90049,4072,161030,2399362,1431417,General VQA
+hateful_memes,8500,8500,8500,128375,17000,General VQA
+iconqa,27307,27307,29841,906877,72492,General VQA
+iconqa(mathv360k),22589,22589,22589,952183,134029,General VQA
+idk,11123,11123,27614,235262,665247,General VQA
+indoor_qa,3350,3350,3350,36832,19700,General VQA
+LLaVA_Instruct_150K,157710,157710,361405,4412600,28719278,General VQA
+llavar_gpt4_20k,19790,19790,43167,546703,1516730,General VQA
+lnqa,302780,302780,1520942,16530323,19027663,General VQA
+lrv_normal(filtered),10489,10489,155269,2108321,3134247,General VQA
+lvis_instruct4v,222711,222711,1050622,12556173,43726782,General VQA
+mimic_cgd,141878,70939,141869,1789740,4304380,General VQA
+MMEvol,160215,160215,630441,16203127,50445237,General VQA
+mmra,2048,1024,1024,72523,25764,General VQA
+nlvr2,100852,50426,86373,4629641,172746,General VQA
+sketchyvqa,8000,8000,8000,182192,8000,General VQA
+spark,3904,3904,6248,65982,73973,General VQA
+spatialsense,10440,10440,17498,200963,418883,General VQA
+spot_the_diff,17132,8566,9524,82670,209630,General VQA
+vision_flan(filtered),175964,175964,175964,9983758,3009891,General VQA
+visual7w,14366,14366,69817,3054334,209451,General VQA
+vizwiz(mathv360k),6604,6604,6604,197143,44876,General VQA
+vqav2,82772,82772,443757,5722488,1100837,General VQA
+vsr,2157,2157,3354,79596,6708,General VQA
+websight,10000,10000,10000,113114,5237381,General VQA
+wildvision,333,333,405,50161,72820,General VQA
+yesbut,4318,4318,4318,38365,157229,General VQA
+aguvis-stage-1,458957,458957,3831666,36151272,93546182,Grounding & Counting
+groundui,13531,13531,18016,200094,883274,Grounding & Counting
+Objects365_QA,1742287,1742287,12329259,135681680,2146619635,Grounding & Counting
+oodvqa,8488,8488,8488,227028,8488,Grounding & Counting
+tallyqa,98680,98680,183986,2674306,370282,Grounding & Counting
+clevr,70000,70000,699989,19277813,1570525,Mathematics
+clevr_math,70000,70000,556082,7888064,580324,Mathematics
+clevr_math(mathv360k),5280,5280,5280,174879,27536,Mathematics
+CoSyn_400k_math,66714,66714,66714,500554,28631388,Mathematics
+geo170k(align),35297,35297,35297,336151,1866019,Mathematics
+geo170k(qa),12101,12101,12101,1254831,1115242,Mathematics
+geo3k,2091,2091,2091,130287,2091,Mathematics
+geometry3k(mathv360k),9724,9724,9724,541908,69075,Mathematics
+geomverse,9303,9303,9339,662756,2454014,Mathematics
+geoqa+(mathv360k),17162,17162,17162,1449094,117740,Mathematics
+geos(mathv360k),498,498,498,32394,3509,Mathematics
+intergps,1280,1280,1760,97799,5280,Mathematics
+mavis_math_metagen,87348,87348,87348,6668920,5486485,Mathematics
+mavis_math_rule_geo,99986,99986,99986,8211079,12535251,Mathematics
+raven,63081,42000,42000,584843,63081,Mathematics
+super_clevr(mathv360k),8642,8642,8642,307438,44129,Mathematics
+unigeo(mathv360k),11949,11949,11949,1011069,81781,Mathematics
+art,5603,5603,5603,56573,283138,Naive OCR
+captcha,113062,113062,113062,1469548,466856,Naive OCR
+chrome_writting,8825,8825,8825,150025,172940,Naive OCR
+cocotext,16169,16169,16169,143818,177111,Naive OCR
+ctw,24290,24290,180621,9787485,1653254,Naive OCR
+funsd,194,194,3879,16856,29996,Naive OCR
+hme100k,74492,74492,74492,1117380,1757743,Naive OCR
+hw_squad,20457,20457,83682,1071534,388518,Naive OCR
+iam,5663,5663,5663,45582,130794,Naive OCR
+iiit5k,1990,1990,1990,35820,4259,Naive OCR
+imgur5k,5934,5934,5934,89010,288054,Naive OCR
+k12_printing,256636,256636,256636,14114980,7465001,Naive OCR
+latex_handwritten,39583,39583,39583,390343,1874733,Naive OCR
+latexformulas,552340,552340,552340,5138603,43094747,Naive OCR
+maptext,200,200,799,9434,70813,Naive OCR
+mathwriting-google,300000,300000,300000,2461270,5954806,Naive OCR
+memotion,6991,6991,6991,194718,177429,Naive OCR
+orand_car_a,1999,1999,1999,43978,9035,Naive OCR
+rendered_text,10000,10000,10000,85879,244183,Naive OCR
+sroie,33616,33616,33616,605088,243240,Naive OCR
+svrd,4396,4396,4396,65400,834514,Naive OCR
+SynthCodeNet,499983,499983,499983,2000683,253422136,Naive OCR
+synthdog,500000,500000,500000,8849848,48010145,Naive OCR
+SynthFormulaNet,499997,499997,499997,1999631,51215097,Naive OCR
+tal_ocr_eng,256646,256646,256646,3385012,7465207,Naive OCR
+wordart,19066,4804,4804,78032,54263,Naive OCR
+olmOCR-mix-0225-documents,228864,228864,228858,2197147,163194337,Naive OCR
+olmOCR-mix-0225-books,15194,15194,15194,145750,7962779,Naive OCR
+a_okvqa,54602,54602,54602,1065188,360990,OCR QA
+aokvqa,16539,16539,17056,743458,218917,OCR QA
+arxivqa,100000,100000,100000,7022001,6422269,OCR QA
+bentham,10843,10843,10843,103042,124459,OCR QA
+blockdiagramcomputerized,502,502,502,5067,34453,OCR QA
+blockdiagramhandwritten,1029,1029,1029,11444,75598,OCR QA
+CoSyn_400k_diagram,34963,34963,300357,3356844,11943321,OCR QA
+CoSyn_400k_document,71282,71282,605173,6216517,16095526,OCR QA
+CoSyn_400k_music,11969,11969,81786,792129,3175586,OCR QA
+CoSyn_400k_nutrition,6931,6931,112097,1642936,3687254,OCR QA
+diagram_image_to_text,300,300,300,3631,20723,OCR QA
+DoclingMatix,2465202,1270911,10626898,162581660,2996338775,OCR QA
+docvqa,10189,10189,39463,724814,275510,OCR QA
+est_vqa,19358,19358,19358,286343,143270,OCR QA
+handwriting_forms,1400,1400,1400,81200,41490,OCR QA
+infographic_vqa,1982,4394,23717,392456,86951,OCR QA
+infographic_vqa_llava_format,4394,2113,10054,174352,43912,OCR QA
+infographic(gpt4v),2113,1982,1982,275498,1044183,OCR QA
+invoices_receipts,3013,3013,3013,36745,771948,OCR QA
+mapqa,37417,37417,483416,8454722,5657339,OCR QA
+mapqa(mathv360k),5225,5225,5225,168390,44560,OCR QA
+mmsoc_memotion,6991,6991,6991,188505,421250,OCR QA
+ocrvqa,165746,165746,801579,12217564,4801833,OCR QA
+pdfvqa,8593,8593,95000,1272618,939948,OCR QA
+screen2words,15730,15730,15743,133014,120781,OCR QA
+screenqa,80761,80761,80761,940729,826795,OCR QA
+slidevqa,11868,1919,10617,333065,156036,OCR QA
+st_vqa,17247,17247,23121,338837,98892,OCR QA
+sujet_finance,9801,9801,107050,1395624,1925361,OCR QA
+textocr(gpt4v),25060,25060,25060,150360,2436974,OCR QA
+textvqa,21953,21953,34602,553990,141882,OCR QA
+ureader_cap,91215,91215,91215,1086484,1435964,OCR QA
+ureader_ie,17320,17320,17320,406237,128229,OCR QA
+ureader_kg_processed,37550,37550,37550,352907,2013731,OCR QA
+ureader_qa_processed,252953,252953,252953,7100750,930617,OCR QA
+visualmrc,3027,3027,11988,139751,147385,OCR QA
+ai2d_merged,4858,4858,12325,755455,1319140,Science
+CoSyn_400k_chemical,8942,8942,55391,634881,2450290,Science
+CoSyn_400k_circuit,10470,10470,67939,713575,2637618,Science
+pathvqa,32632,32632,32632,639385,85168,Science
+pmc_vqa(mathv360k),35948,35948,35948,1889167,255109,Science
+scienceqa,4976,4976,6149,1081220,18447,Science
+scienceqa(nona_context),19208,19208,19208,1624583,25311,Science
+tqa,2749,2749,12567,395956,149776,Science
+visualwebinstruct(filtered),263581,263581,263581,8341540,31802459,Science
+vqarad,313,313,1793,25181,6003,Science
+text_code_feedback,0,66383,221096,19349056,79752351,Text-only
+text_codefeedback_filtered_instruction,0,156525,156525,27684170,62764414,Text-only
+text_infinitymath,0,101380,101380,9158132,212543,Text-only
+text_mathinstruct,0,262039,262039,20405295,44145362,Text-only
+text_mathqa,0,394996,394996,23552035,72451061,Text-only
+text_mathstepdpo10k,0,10795,10795,557233,989312,Text-only
+text_numinamath_cot,0,859494,859494,75818870,387758581,Text-only
+text_openhermes_2_5,0,1001551,1008268,142376960,233561291,Text-only
+text_openorca,0,4233853,4233853,1049478873,468042176,Text-only
+text_orcamath,0,200035,200035,12691014,61860987,Text-only
+text_pythoncode25k,0,49626,49626,1629286,4945892,Text-only
+text_pythoncodealpaca,0,18612,18612,655127,2683469,Text-only
+text_ruozhiba,0,1496,1496,69795,234822,Text-only
+text_theoremqa,0,800,800,50065,3468,Text-only
+text_wizardlm_evol,0,69999,69999,7753963,21955856,Text-only
+text_OpenMathInstruct-2,0,1000000,1000000,74905850,413132418,Text-only