{
  "train_key": "faca7844a506b4f7",
  "finetuned": {
    "all": {
      "n": 101716,
      "accuracy": 0.6859,
      "brier": 0.4097,
      "nll": 0.8512,
      "ece": 0.0834
    },
    "source/bev_tr": {
      "n": 46320,
      "accuracy": 0.7745,
      "brier": 0.3074,
      "nll": 0.5773,
      "ece": 0.0489
    },
    "source/bev_tr/choice": {
      "n": 18808,
      "accuracy": 0.7425,
      "brier": 0.3398,
      "nll": 0.6614,
      "ece": 0.0608
    },
    "source/bev_tr/noul": {
      "n": 17572,
      "accuracy": 0.8753,
      "brier": 0.1909,
      "nll": 0.3247,
      "ece": 0.0429
    },
    "source/bev_tr/score": {
      "n": 9940,
      "accuracy": 0.6568,
      "brier": 0.4523,
      "nll": 0.8645,
      "ece": 0.069
    },
    "source/bev_tr_en": {
      "n": 5766,
      "accuracy": 0.7718,
      "brier": 0.3069,
      "nll": 0.5821,
      "ece": 0.0683
    },
    "source/bev_tr_en/choice": {
      "n": 2386,
      "accuracy": 0.7297,
      "brier": 0.3496,
      "nll": 0.6823,
      "ece": 0.07
    },
    "source/bev_tr_en/noul": {
      "n": 2117,
      "accuracy": 0.8857,
      "brier": 0.1741,
      "nll": 0.3005,
      "ece": 0.0652
    },
    "source/bev_tr_en/score": {
      "n": 1263,
      "accuracy": 0.6603,
      "brier": 0.449,
      "nll": 0.8647,
      "ece": 0.0859
    },
    "source/beyazperde_tr": {
      "n": 5000,
      "accuracy": 0.5334,
      "brier": 0.6079,
      "nll": 1.1354,
      "ece": 0.1193
    },
    "source/beyazperde_tr/score": {
      "n": 5000,
      "accuracy": 0.5334,
      "brier": 0.6079,
      "nll": 1.1354,
      "ece": 0.1193
    },
    "source/jev_bench_tr": {
      "n": 22773,
      "accuracy": 0.4957,
      "brier": 0.5956,
      "nll": 1.509,
      "ece": 0.1748
    },
    "source/jev_bench_tr/choice": {
      "n": 9599,
      "accuracy": 0.4197,
      "brier": 0.6558,
      "nll": 2.1582,
      "ece": 0.1971
    },
    "source/jev_bench_tr/noul": {
      "n": 7174,
      "accuracy": 0.7346,
      "brier": 0.3753,
      "nll": 0.6143,
      "ece": 0.1118
    },
    "source/jev_bench_tr/score": {
      "n": 6000,
      "accuracy": 0.3317,
      "brier": 0.7626,
      "nll": 1.5402,
      "ece": 0.2147
    },
    "source/jev_bench_tr_en": {
      "n": 2000,
      "accuracy": 0.5835,
      "brier": 0.4941,
      "nll": 1.2463,
      "ece": 0.1094
    },
    "source/jev_bench_tr_en/choice": {
      "n": 839,
      "accuracy": 0.5185,
      "brier": 0.5398,
      "nll": 1.7566,
      "ece": 0.1387
    },
    "source/jev_bench_tr_en/noul": {
      "n": 635,
      "accuracy": 0.8094,
      "brier": 0.2705,
      "nll": 0.4705,
      "ece": 0.0497
    },
    "source/jev_bench_tr_en/score": {
      "n": 526,
      "accuracy": 0.4144,
      "brier": 0.691,
      "nll": 1.369,
      "ece": 0.162
    },
    "source/massive_tr": {
      "n": 2974,
      "accuracy": 0.773,
      "brier": 0.3361,
      "nll": 0.9659,
      "ece": 0.069
    },
    "source/massive_tr/choice": {
      "n": 2974,
      "accuracy": 0.773,
      "brier": 0.3361,
      "nll": 0.9659,
      "ece": 0.069
    },
    "source/squad_tr": {
      "n": 11873,
      "accuracy": 0.7028,
      "brier": 0.4297,
      "nll": 0.6379,
      "ece": 0.1329
    },
    "source/squad_tr/noul": {
      "n": 11873,
      "accuracy": 0.7028,
      "brier": 0.4297,
      "nll": 0.6379,
      "ece": 0.1329
    },
    "source/xnli_tr": {
      "n": 5010,
      "accuracy": 0.7329,
      "brier": 0.3943,
      "nll": 0.6994,
      "ece": 0.0974
    },
    "source/xnli_tr/choice": {
      "n": 5010,
      "accuracy": 0.7329,
      "brier": 0.3943,
      "nll": 0.6994,
      "ece": 0.0974
    },
    "type/choice": {
      "n": 39616,
      "accuracy": 0.6599,
      "brier": 0.4278,
      "nll": 1.0762,
      "ece": 0.0975
    },
    "type/noul": {
      "n": 39371,
      "accuracy": 0.7971,
      "brier": 0.2969,
      "nll": 0.473,
      "ece": 0.0598
    },
    "type/score": {
      "n": 22729,
      "accuracy": 0.5384,
      "brier": 0.5738,
      "nll": 1.1142,
      "ece": 0.108
    },
    "workflow/bev_tr/Browser interaction": {
      "n": 286,
      "accuracy": 0.7238,
      "brier": 0.4303,
      "nll": 0.8399,
      "ece": 0.1325
    },
    "workflow/bev_tr/Civic and safety operations": {
      "n": 261,
      "accuracy": 0.6284,
      "brier": 0.4048,
      "nll": 0.8256,
      "ece": 0.092
    },
    "workflow/bev_tr/Contract evidence": {
      "n": 320,
      "accuracy": 0.725,
      "brier": 0.3853,
      "nll": 0.6728,
      "ece": 0.0802
    },
    "workflow/bev_tr/Engagement and ranking": {
      "n": 82,
      "accuracy": 0.622,
      "brier": 0.5174,
      "nll": 0.7323,
      "ece": 0.2016
    },
    "workflow/bev_tr/Financial reporting and banking": {
      "n": 1740,
      "accuracy": 0.6034,
      "brier": 0.5056,
      "nll": 0.9542,
      "ece": 0.0314
    },
    "workflow/bev_tr/Game-state decisions": {
      "n": 2039,
      "accuracy": 0.8583,
      "brier": 0.2005,
      "nll": 0.3516,
      "ece": 0.0662
    },
    "workflow/bev_tr/Reading comprehension": {
      "n": 1422,
      "accuracy": 0.7124,
      "brier": 0.4185,
      "nll": 0.6232,
      "ece": 0.1233
    },
    "workflow/bev_tr/Response preference and quality": {
      "n": 3176,
      "accuracy": 0.5608,
      "brier": 0.564,
      "nll": 1.0111,
      "ece": 0.0713
    },
    "workflow/bev_tr/Retail, product, and shopping": {
      "n": 6601,
      "accuracy": 0.9294,
      "brier": 0.1102,
      "nll": 0.2472,
      "ece": 0.0669
    },
    "workflow/bev_tr/Scientific and paper understanding": {
      "n": 508,
      "accuracy": 0.8484,
      "brier": 0.2478,
      "nll": 0.4493,
      "ece": 0.058
    },
    "workflow/bev_tr/Sensory quality rating": {
      "n": 260,
      "accuracy": 0.4538,
      "brier": 0.6853,
      "nll": 1.3114,
      "ece": 0.0915
    },
    "workflow/bev_tr/Sentiment, emotion, and moderation": {
      "n": 11526,
      "accuracy": 0.8601,
      "brier": 0.215,
      "nll": 0.4164,
      "ece": 0.0521
    },
    "workflow/bev_tr/Software engineering and code": {
      "n": 2039,
      "accuracy": 0.6077,
      "brier": 0.5418,
      "nll": 1.0388,
      "ece": 0.1409
    },
    "workflow/bev_tr/Software security": {
      "n": 2240,
      "accuracy": 0.6661,
      "brier": 0.4489,
      "nll": 0.7744,
      "ece": 0.0795
    },
    "workflow/bev_tr/Spam detection": {
      "n": 271,
      "accuracy": 0.9557,
      "brier": 0.0837,
      "nll": 0.1827,
      "ece": 0.079
    },
    "workflow/bev_tr/Spatial and logical reasoning": {
      "n": 8384,
      "accuracy": 0.5964,
      "brier": 0.5129,
      "nll": 0.9455,
      "ece": 0.0634
    },
    "workflow/bev_tr/Support and intent routing": {
      "n": 2858,
      "accuracy": 0.9689,
      "brier": 0.0529,
      "nll": 0.1514,
      "ece": 0.0532
    },
    "workflow/bev_tr/Tool and workflow decisions": {
      "n": 2307,
      "accuracy": 0.984,
      "brier": 0.0335,
      "nll": 0.1172,
      "ece": 0.0626
    },
    "workflow/bev_tr_en/Browser interaction": {
      "n": 39,
      "accuracy": 0.5385,
      "brier": 0.5845,
      "nll": 1.109,
      "ece": 0.2958
    },
    "workflow/bev_tr_en/Civic and safety operations": {
      "n": 31,
      "accuracy": 0.4839,
      "brier": 0.5642,
      "nll": 1.1315,
      "ece": 0.1902
    },
    "workflow/bev_tr_en/Contract evidence": {
      "n": 49,
      "accuracy": 0.7551,
      "brier": 0.342,
      "nll": 0.6166,
      "ece": 0.1222
    },
    "workflow/bev_tr_en/Engagement and ranking": {
      "n": 6,
      "accuracy": 0.6667,
      "brier": 0.4493,
      "nll": 0.6854,
      "ece": 0.358
    },
    "workflow/bev_tr_en/Financial reporting and banking": {
      "n": 196,
      "accuracy": 0.6735,
      "brier": 0.4468,
      "nll": 0.9134,
      "ece": 0.0699
    },
    "workflow/bev_tr_en/Game-state decisions": {
      "n": 303,
      "accuracy": 0.8746,
      "brier": 0.1858,
      "nll": 0.3338,
      "ece": 0.0784
    },
    "workflow/bev_tr_en/Reading comprehension": {
      "n": 168,
      "accuracy": 0.6964,
      "brier": 0.4399,
      "nll": 0.6425,
      "ece": 0.1711
    },
    "workflow/bev_tr_en/Response preference and quality": {
      "n": 389,
      "accuracy": 0.5064,
      "brier": 0.6134,
      "nll": 1.1173,
      "ece": 0.124
    },
    "workflow/bev_tr_en/Retail, product, and shopping": {
      "n": 902,
      "accuracy": 0.9335,
      "brier": 0.1057,
      "nll": 0.2486,
      "ece": 0.0839
    },
    "workflow/bev_tr_en/Scientific and paper understanding": {
      "n": 66,
      "accuracy": 0.8333,
      "brier": 0.2617,
      "nll": 0.4858,
      "ece": 0.0454
    },
    "workflow/bev_tr_en/Sensory quality rating": {
      "n": 37,
      "accuracy": 0.4054,
      "brier": 0.7053,
      "nll": 1.3465,
      "ece": 0.1369
    },
    "workflow/bev_tr_en/Sentiment, emotion, and moderation": {
      "n": 1439,
      "accuracy": 0.8902,
      "brier": 0.172,
      "nll": 0.3386,
      "ece": 0.0635
    },
    "workflow/bev_tr_en/Software engineering and code": {
      "n": 260,
      "accuracy": 0.6,
      "brier": 0.5622,
      "nll": 1.0571,
      "ece": 0.1826
    },
    "workflow/bev_tr_en/Software security": {
      "n": 234,
      "accuracy": 0.6197,
      "brier": 0.4952,
      "nll": 0.8507,
      "ece": 0.138
    },
    "workflow/bev_tr_en/Spam detection": {
      "n": 23,
      "accuracy": 1.0,
      "brier": 0.0144,
      "nll": 0.0881,
      "ece": 0.0842
    },
    "workflow/bev_tr_en/Spatial and logical reasoning": {
      "n": 1042,
      "accuracy": 0.5509,
      "brier": 0.5594,
      "nll": 1.0338,
      "ece": 0.0919
    },
    "workflow/bev_tr_en/Support and intent routing": {
      "n": 310,
      "accuracy": 0.971,
      "brier": 0.0545,
      "nll": 0.1612,
      "ece": 0.0499
    },
    "workflow/bev_tr_en/Tool and workflow decisions": {
      "n": 272,
      "accuracy": 0.9926,
      "brier": 0.0197,
      "nll": 0.0891,
      "ece": 0.0748
    },
    "workflow/beyazperde_tr": {
      "n": 5000,
      "accuracy": 0.5334,
      "brier": 0.6079,
      "nll": 1.1354,
      "ece": 0.1193
    },
    "workflow/jev_bench_tr/arc_challenge": {
      "n": 1000,
      "accuracy": 0.249,
      "brier": 0.9511,
      "nll": 1.8115,
      "ece": 0.3312
    },
    "workflow/jev_bench_tr/banking77": {
      "n": 1000,
      "accuracy": 0.319,
      "brier": 0.9801,
      "nll": 3.3462,
      "ece": 0.3322
    },
    "workflow/jev_bench_tr/boolq": {
      "n": 1000,
      "accuracy": 0.711,
      "brier": 0.4183,
      "nll": 0.6259,
      "ece": 0.1223
    },
    "workflow/jev_bench_tr/chaosnli": {
      "n": 1599,
      "accuracy": 0.546,
      "brier": 0.3285,
      "nll": 1.3672,
      "ece": 0.252
    },
    "workflow/jev_bench_tr/civil_comments": {
      "n": 2000,
      "accuracy": 0.8565,
      "brier": 0.143,
      "nll": 0.4014,
      "ece": 0.0298
    },
    "workflow/jev_bench_tr/clinc150": {
      "n": 1000,
      "accuracy": 0.39,
      "brier": 0.7676,
      "nll": 2.761,
      "ece": 0.0503
    },
    "workflow/jev_bench_tr/fever_evidence": {
      "n": 1000,
      "accuracy": 0.813,
      "brier": 0.2784,
      "nll": 0.4417,
      "ece": 0.0277
    },
    "workflow/jev_bench_tr/go_emotions": {
      "n": 1000,
      "accuracy": 0.307,
      "brier": 0.4861,
      "nll": 3.0264,
      "ece": 0.2539
    },
    "workflow/jev_bench_tr/helpsteer2_helpfulness": {
      "n": 1000,
      "accuracy": 0.164,
      "brier": 0.908,
      "nll": 1.8281,
      "ece": 0.2921
    },
    "workflow/jev_bench_tr/helpsteer2_verbosity": {
      "n": 1000,
      "accuracy": 0.214,
      "brier": 0.8105,
      "nll": 1.4787,
      "ece": 0.2734
    },
    "workflow/jev_bench_tr/ledgar": {
      "n": 1000,
      "accuracy": 0.092,
      "brier": 0.9895,
      "nll": 4.2557,
      "ece": 0.1183
    },
    "workflow/jev_bench_tr/massive": {
      "n": 1000,
      "accuracy": 0.775,
      "brier": 0.338,
      "nll": 0.9897,
      "ece": 0.0666
    },
    "workflow/jev_bench_tr/measuring_hate_speech": {
      "n": 1000,
      "accuracy": 0.563,
      "brier": 0.48,
      "nll": 1.2224,
      "ece": 0.202
    },
    "workflow/jev_bench_tr/mmlu": {
      "n": 1000,
      "accuracy": 0.267,
      "brier": 0.89,
      "nll": 1.6806,
      "ece": 0.2749
    },
    "workflow/jev_bench_tr/mnli": {
      "n": 1000,
      "accuracy": 0.757,
      "brier": 0.3668,
      "nll": 0.6591,
      "ece": 0.0882
    },
    "workflow/jev_bench_tr/paws": {
      "n": 1000,
      "accuracy": 0.541,
      "brier": 0.692,
      "nll": 1.0228,
      "ece": 0.3053
    },
    "workflow/jev_bench_tr/sms_spam": {
      "n": 800,
      "accuracy": 0.9563,
      "brier": 0.0793,
      "nll": 0.1781,
      "ece": 0.0789
    },
    "workflow/jev_bench_tr/sst5": {
      "n": 1000,
      "accuracy": 0.385,
      "brier": 0.7825,
      "nll": 1.5274,
      "ece": 0.2169
    },
    "workflow/jev_bench_tr/strategyqa_closed": {
      "n": 687,
      "accuracy": 0.5109,
      "brier": 0.7302,
      "nll": 1.0431,
      "ece": 0.316
    },
    "workflow/jev_bench_tr/strategyqa_grounded": {
      "n": 687,
      "accuracy": 0.5473,
      "brier": 0.6591,
      "nll": 0.953,
      "ece": 0.2686
    },
    "workflow/jev_bench_tr/stsb": {
      "n": 1000,
      "accuracy": 0.242,
      "brier": 0.8549,
      "nll": 1.7831,
      "ece": 0.1647
    },
    "workflow/jev_bench_tr/yelp5": {
      "n": 1000,
      "accuracy": 0.422,
      "brier": 0.7397,
      "nll": 1.4017,
      "ece": 0.1902
    },
    "workflow/jev_bench_tr_en/arc_challenge": {
      "n": 91,
      "accuracy": 0.2857,
      "brier": 0.9021,
      "nll": 1.7181,
      "ece": 0.3246
    },
    "workflow/jev_bench_tr_en/banking77": {
      "n": 71,
      "accuracy": 0.3803,
      "brier": 0.8703,
      "nll": 2.9173,
      "ece": 0.3313
    },
    "workflow/jev_bench_tr_en/boolq": {
      "n": 80,
      "accuracy": 0.7,
      "brier": 0.4287,
      "nll": 0.6398,
      "ece": 0.1632
    },
    "workflow/jev_bench_tr_en/chaosnli": {
      "n": 151,
      "accuracy": 0.6358,
      "brier": 0.2489,
      "nll": 1.2183,
      "ece": 0.1784
    },
    "workflow/jev_bench_tr_en/civil_comments": {
      "n": 171,
      "accuracy": 0.9006,
      "brier": 0.0917,
      "nll": 0.3407,
      "ece": 0.0876
    },
    "workflow/jev_bench_tr_en/clinc150": {
      "n": 96,
      "accuracy": 0.5938,
      "brier": 0.5424,
      "nll": 2.0461,
      "ece": 0.0991
    },
    "workflow/jev_bench_tr_en/fever_evidence": {
      "n": 91,
      "accuracy": 0.9011,
      "brier": 0.1923,
      "nll": 0.3367,
      "ece": 0.1187
    },
    "workflow/jev_bench_tr_en/go_emotions": {
      "n": 93,
      "accuracy": 0.4409,
      "brier": 0.3723,
      "nll": 2.7647,
      "ece": 0.1492
    },
    "workflow/jev_bench_tr_en/helpsteer2_helpfulness": {
      "n": 89,
      "accuracy": 0.2472,
      "brier": 0.8271,
      "nll": 1.6234,
      "ece": 0.1724
    },
    "workflow/jev_bench_tr_en/helpsteer2_verbosity": {
      "n": 79,
      "accuracy": 0.3038,
      "brier": 0.7319,
      "nll": 1.3641,
      "ece": 0.0899
    },
    "workflow/jev_bench_tr_en/ledgar": {
      "n": 70,
      "accuracy": 0.3,
      "brier": 0.8188,
      "nll": 2.8736,
      "ece": 0.1541
    },
    "workflow/jev_bench_tr_en/massive": {
      "n": 95,
      "accuracy": 0.6947,
      "brier": 0.4233,
      "nll": 1.165,
      "ece": 0.0782
    },
    "workflow/jev_bench_tr_en/measuring_hate_speech": {
      "n": 96,
      "accuracy": 0.6042,
      "brier": 0.4065,
      "nll": 1.0435,
      "ece": 0.179
    },
    "workflow/jev_bench_tr_en/mmlu": {
      "n": 80,
      "accuracy": 0.2875,
      "brier": 0.8719,
      "nll": 1.6309,
      "ece": 0.3092
    },
    "workflow/jev_bench_tr_en/mnli": {
      "n": 92,
      "accuracy": 0.8478,
      "brier": 0.1896,
      "nll": 0.3318,
      "ece": 0.1131
    },
    "workflow/jev_bench_tr_en/paws": {
      "n": 80,
      "accuracy": 0.8,
      "brier": 0.3218,
      "nll": 0.4965,
      "ece": 0.0879
    },
    "workflow/jev_bench_tr_en/sms_spam": {
      "n": 82,
      "accuracy": 0.9634,
      "brier": 0.0574,
      "nll": 0.1469,
      "ece": 0.0827
    },
    "workflow/jev_bench_tr_en/sst5": {
      "n": 73,
      "accuracy": 0.4521,
      "brier": 0.7319,
      "nll": 1.3825,
      "ece": 0.2967
    },
    "workflow/jev_bench_tr_en/strategyqa_closed": {
      "n": 60,
      "accuracy": 0.55,
      "brier": 0.5863,
      "nll": 0.8265,
      "ece": 0.2422
    },
    "workflow/jev_bench_tr_en/strategyqa_grounded": {
      "n": 71,
      "accuracy": 0.6479,
      "brier": 0.5446,
      "nll": 0.8072,
      "ece": 0.2425
    },
    "workflow/jev_bench_tr_en/stsb": {
      "n": 90,
      "accuracy": 0.2889,
      "brier": 0.8276,
      "nll": 1.7023,
      "ece": 0.1884
    },
    "workflow/jev_bench_tr_en/yelp5": {
      "n": 99,
      "accuracy": 0.5556,
      "brier": 0.6575,
      "nll": 1.1468,
      "ece": 0.222
    },
    "workflow/massive_tr": {
      "n": 2974,
      "accuracy": 0.773,
      "brier": 0.3361,
      "nll": 0.9659,
      "ece": 0.069
    },
    "workflow/squad_tr": {
      "n": 11873,
      "accuracy": 0.7028,
      "brier": 0.4297,
      "nll": 0.6379,
      "ece": 0.1329
    },
    "workflow/xnli_tr": {
      "n": 5010,
      "accuracy": 0.7329,
      "brier": 0.3943,
      "nll": 0.6994,
      "ece": 0.0974
    }
  },
  "base": {
    "all": {
      "n": 101716,
      "accuracy": 0.4425,
      "brier": 0.7556,
      "nll": 1.9943,
      "ece": 0.282
    },
    "source/bev_tr": {
      "n": 46320,
      "accuracy": 0.4437,
      "brier": 0.7622,
      "nll": 2.0489,
      "ece": 0.2748
    },
    "source/bev_tr/choice": {
      "n": 18808,
      "accuracy": 0.4496,
      "brier": 0.7135,
      "nll": 1.574,
      "ece": 0.2179
    },
    "source/bev_tr/noul": {
      "n": 17572,
      "accuracy": 0.5741,
      "brier": 0.6909,
      "nll": 2.5173,
      "ece": 0.2977
    },
    "source/bev_tr/score": {
      "n": 9940,
      "accuracy": 0.2022,
      "brier": 0.9806,
      "nll": 2.1193,
      "ece": 0.342
    },
    "source/bev_tr_en": {
      "n": 5766,
      "accuracy": 0.4728,
      "brier": 0.7083,
      "nll": 1.8292,
      "ece": 0.2426
    },
    "source/bev_tr_en/choice": {
      "n": 2386,
      "accuracy": 0.4593,
      "brier": 0.6881,
      "nll": 1.5259,
      "ece": 0.2123
    },
    "source/bev_tr_en/noul": {
      "n": 2117,
      "accuracy": 0.633,
      "brier": 0.5712,
      "nll": 1.8677,
      "ece": 0.2396
    },
    "source/bev_tr_en/score": {
      "n": 1263,
      "accuracy": 0.2296,
      "brier": 0.9762,
      "nll": 2.3378,
      "ece": 0.3121
    },
    "source/beyazperde_tr": {
      "n": 5000,
      "accuracy": 0.1796,
      "brier": 0.9537,
      "nll": 1.9932,
      "ece": 0.3297
    },
    "source/beyazperde_tr/score": {
      "n": 5000,
      "accuracy": 0.1796,
      "brier": 0.9537,
      "nll": 1.9932,
      "ece": 0.3297
    },
    "source/jev_bench_tr": {
      "n": 22773,
      "accuracy": 0.4019,
      "brier": 0.7689,
      "nll": 2.3281,
      "ece": 0.2951
    },
    "source/jev_bench_tr/choice": {
      "n": 9599,
      "accuracy": 0.3403,
      "brier": 0.8114,
      "nll": 3.1438,
      "ece": 0.3181
    },
    "source/jev_bench_tr/noul": {
      "n": 7174,
      "accuracy": 0.6387,
      "brier": 0.5951,
      "nll": 1.4891,
      "ece": 0.2638
    },
    "source/jev_bench_tr/score": {
      "n": 6000,
      "accuracy": 0.2172,
      "brier": 0.9088,
      "nll": 2.0262,
      "ece": 0.2957
    },
    "source/jev_bench_tr_en": {
      "n": 2000,
      "accuracy": 0.4995,
      "brier": 0.6503,
      "nll": 2.0143,
      "ece": 0.2461
    },
    "source/jev_bench_tr_en/choice": {
      "n": 839,
      "accuracy": 0.4803,
      "brier": 0.6585,
      "nll": 2.6276,
      "ece": 0.257
    },
    "source/jev_bench_tr_en/noul": {
      "n": 635,
      "accuracy": 0.7213,
      "brier": 0.4066,
      "nll": 0.9011,
      "ece": 0.1574
    },
    "source/jev_bench_tr_en/score": {
      "n": 526,
      "accuracy": 0.2624,
      "brier": 0.9312,
      "nll": 2.38,
      "ece": 0.3379
    },
    "source/massive_tr": {
      "n": 2974,
      "accuracy": 0.3534,
      "brier": 0.8701,
      "nll": 3.228,
      "ece": 0.2862
    },
    "source/massive_tr/choice": {
      "n": 2974,
      "accuracy": 0.3534,
      "brier": 0.8701,
      "nll": 3.228,
      "ece": 0.2862
    },
    "source/squad_tr": {
      "n": 11873,
      "accuracy": 0.5198,
      "brier": 0.7361,
      "nll": 1.2571,
      "ece": 0.319
    },
    "source/squad_tr/noul": {
      "n": 11873,
      "accuracy": 0.5198,
      "brier": 0.7361,
      "nll": 1.2571,
      "ece": 0.319
    },
    "source/xnli_tr": {
      "n": 5010,
      "accuracy": 0.6912,
      "brier": 0.5108,
      "nll": 1.1705,
      "ece": 0.2123
    },
    "source/xnli_tr/choice": {
      "n": 5010,
      "accuracy": 0.6912,
      "brier": 0.5108,
      "nll": 1.1705,
      "ece": 0.2123
    },
    "type/choice": {
      "n": 39616,
      "accuracy": 0.4477,
      "brier": 0.7206,
      "nll": 2.0469,
      "ece": 0.247
    },
    "type/noul": {
      "n": 39371,
      "accuracy": 0.575,
      "brier": 0.6761,
      "nll": 1.8889,
      "ece": 0.2923
    },
    "type/score": {
      "n": 22729,
      "accuracy": 0.2041,
      "brier": 0.9543,
      "nll": 2.0851,
      "ece": 0.3253
    },
    "workflow/bev_tr/Browser interaction": {
      "n": 286,
      "accuracy": 0.2727,
      "brier": 0.8967,
      "nll": 1.9323,
      "ece": 0.2102
    },
    "workflow/bev_tr/Civic and safety operations": {
      "n": 261,
      "accuracy": 0.2529,
      "brier": 0.9124,
      "nll": 1.996,
      "ece": 0.2477
    },
    "workflow/bev_tr/Contract evidence": {
      "n": 320,
      "accuracy": 0.3344,
      "brier": 0.9895,
      "nll": 1.7099,
      "ece": 0.4493
    },
    "workflow/bev_tr/Engagement and ranking": {
      "n": 82,
      "accuracy": 0.5244,
      "brier": 0.5376,
      "nll": 0.7463,
      "ece": 0.1505
    },
    "workflow/bev_tr/Financial reporting and banking": {
      "n": 1740,
      "accuracy": 0.1661,
      "brier": 1.2773,
      "nll": 8.081,
      "ece": 0.5405
    },
    "workflow/bev_tr/Game-state decisions": {
      "n": 2039,
      "accuracy": 0.3453,
      "brier": 0.9941,
      "nll": 2.8814,
      "ece": 0.4135
    },
    "workflow/bev_tr/Reading comprehension": {
      "n": 1422,
      "accuracy": 0.628,
      "brier": 0.5811,
      "nll": 1.0833,
      "ece": 0.2487
    },
    "workflow/bev_tr/Response preference and quality": {
      "n": 3176,
      "accuracy": 0.3212,
      "brier": 0.8784,
      "nll": 1.9156,
      "ece": 0.2694
    },
    "workflow/bev_tr/Retail, product, and shopping": {
      "n": 6601,
      "accuracy": 0.2881,
      "brier": 1.0247,
      "nll": 3.8517,
      "ece": 0.4399
    },
    "workflow/bev_tr/Scientific and paper understanding": {
      "n": 508,
      "accuracy": 0.374,
      "brier": 0.9243,
      "nll": 1.8104,
      "ece": 0.4008
    },
    "workflow/bev_tr/Sensory quality rating": {
      "n": 260,
      "accuracy": 0.0731,
      "brier": 1.105,
      "nll": 2.5661,
      "ece": 0.4819
    },
    "workflow/bev_tr/Sentiment, emotion, and moderation": {
      "n": 11526,
      "accuracy": 0.6159,
      "brier": 0.4961,
      "nll": 0.896,
      "ece": 0.1553
    },
    "workflow/bev_tr/Software engineering and code": {
      "n": 2039,
      "accuracy": 0.3217,
      "brier": 0.7934,
      "nll": 1.502,
      "ece": 0.1479
    },
    "workflow/bev_tr/Software security": {
      "n": 2240,
      "accuracy": 0.4174,
      "brier": 0.856,
      "nll": 1.88,
      "ece": 0.3529
    },
    "workflow/bev_tr/Spam detection": {
      "n": 271,
      "accuracy": 0.5498,
      "brier": 0.7615,
      "nll": 2.0809,
      "ece": 0.3966
    },
    "workflow/bev_tr/Spatial and logical reasoning": {
      "n": 8384,
      "accuracy": 0.3211,
      "brier": 0.8962,
      "nll": 1.9157,
      "ece": 0.3166
    },
    "workflow/bev_tr/Support and intent routing": {
      "n": 2858,
      "accuracy": 0.6777,
      "brier": 0.4581,
      "nll": 0.9635,
      "ece": 0.1267
    },
    "workflow/bev_tr/Tool and workflow decisions": {
      "n": 2307,
      "accuracy": 0.7694,
      "brier": 0.3385,
      "nll": 0.7234,
      "ece": 0.1086
    },
    "workflow/bev_tr_en/Browser interaction": {
      "n": 39,
      "accuracy": 0.2821,
      "brier": 0.9332,
      "nll": 2.1859,
      "ece": 0.3528
    },
    "workflow/bev_tr_en/Civic and safety operations": {
      "n": 31,
      "accuracy": 0.129,
      "brier": 1.051,
      "nll": 2.4101,
      "ece": 0.4106
    },
    "workflow/bev_tr_en/Contract evidence": {
      "n": 49,
      "accuracy": 0.4286,
      "brier": 0.8487,
      "nll": 1.4637,
      "ece": 0.4014
    },
    "workflow/bev_tr_en/Engagement and ranking": {
      "n": 6,
      "accuracy": 0.6667,
      "brier": 0.4494,
      "nll": 0.6345,
      "ece": 0.2668
    },
    "workflow/bev_tr_en/Financial reporting and banking": {
      "n": 196,
      "accuracy": 0.1429,
      "brier": 1.2968,
      "nll": 7.8204,
      "ece": 0.5675
    },
    "workflow/bev_tr_en/Game-state decisions": {
      "n": 303,
      "accuracy": 0.5017,
      "brier": 0.6864,
      "nll": 1.193,
      "ece": 0.1938
    },
    "workflow/bev_tr_en/Reading comprehension": {
      "n": 168,
      "accuracy": 0.6429,
      "brier": 0.5117,
      "nll": 0.8719,
      "ece": 0.2299
    },
    "workflow/bev_tr_en/Response preference and quality": {
      "n": 389,
      "accuracy": 0.3316,
      "brier": 0.85,
      "nll": 1.7891,
      "ece": 0.2496
    },
    "workflow/bev_tr_en/Retail, product, and shopping": {
      "n": 902,
      "accuracy": 0.2993,
      "brier": 0.9607,
      "nll": 3.1195,
      "ece": 0.3775
    },
    "workflow/bev_tr_en/Scientific and paper understanding": {
      "n": 66,
      "accuracy": 0.3485,
      "brier": 1.0125,
      "nll": 2.1613,
      "ece": 0.4552
    },
    "workflow/bev_tr_en/Sensory quality rating": {
      "n": 37,
      "accuracy": 0.1081,
      "brier": 0.8938,
      "nll": 1.8512,
      "ece": 0.2324
    },
    "workflow/bev_tr_en/Sentiment, emotion, and moderation": {
      "n": 1439,
      "accuracy": 0.7137,
      "brier": 0.4125,
      "nll": 0.9542,
      "ece": 0.1372
    },
    "workflow/bev_tr_en/Software engineering and code": {
      "n": 260,
      "accuracy": 0.3192,
      "brier": 0.7658,
      "nll": 1.4263,
      "ece": 0.1422
    },
    "workflow/bev_tr_en/Software security": {
      "n": 234,
      "accuracy": 0.3718,
      "brier": 0.8644,
      "nll": 1.72,
      "ece": 0.3093
    },
    "workflow/bev_tr_en/Spam detection": {
      "n": 23,
      "accuracy": 0.5652,
      "brier": 0.6466,
      "nll": 1.6831,
      "ece": 0.3296
    },
    "workflow/bev_tr_en/Spatial and logical reasoning": {
      "n": 1042,
      "accuracy": 0.3081,
      "brier": 0.8791,
      "nll": 1.8543,
      "ece": 0.2863
    },
    "workflow/bev_tr_en/Support and intent routing": {
      "n": 310,
      "accuracy": 0.7194,
      "brier": 0.3894,
      "nll": 0.842,
      "ece": 0.1377
    },
    "workflow/bev_tr_en/Tool and workflow decisions": {
      "n": 272,
      "accuracy": 0.8015,
      "brier": 0.29,
      "nll": 0.6316,
      "ece": 0.1362
    },
    "workflow/beyazperde_tr": {
      "n": 5000,
      "accuracy": 0.1796,
      "brier": 0.9537,
      "nll": 1.9932,
      "ece": 0.3297
    },
    "workflow/jev_bench_tr/arc_challenge": {
      "n": 1000,
      "accuracy": 0.251,
      "brier": 0.8757,
      "nll": 1.666,
      "ece": 0.2373
    },
    "workflow/jev_bench_tr/banking77": {
      "n": 1000,
      "accuracy": 0.27,
      "brier": 1.1486,
      "nll": 5.3545,
      "ece": 0.4698
    },
    "workflow/jev_bench_tr/boolq": {
      "n": 1000,
      "accuracy": 0.687,
      "brier": 0.5096,
      "nll": 1.0109,
      "ece": 0.2271
    },
    "workflow/jev_bench_tr/chaosnli": {
      "n": 1599,
      "accuracy": 0.4778,
      "brier": 0.4821,
      "nll": 2.3189,
      "ece": 0.3986
    },
    "workflow/jev_bench_tr/civil_comments": {
      "n": 2000,
      "accuracy": 0.9135,
      "brier": 0.0821,
      "nll": 0.3281,
      "ece": 0.0233
    },
    "workflow/jev_bench_tr/clinc150": {
      "n": 1000,
      "accuracy": 0.249,
      "brier": 0.9825,
      "nll": 4.0166,
      "ece": 0.2863
    },
    "workflow/jev_bench_tr/fever_evidence": {
      "n": 1000,
      "accuracy": 0.551,
      "brier": 0.7446,
      "nll": 1.4529,
      "ece": 0.3501
    },
    "workflow/jev_bench_tr/go_emotions": {
      "n": 1000,
      "accuracy": 0.277,
      "brier": 0.7296,
      "nll": 4.5711,
      "ece": 0.4809
    },
    "workflow/jev_bench_tr/helpsteer2_helpfulness": {
      "n": 1000,
      "accuracy": 0.148,
      "brier": 0.9073,
      "nll": 1.896,
      "ece": 0.2568
    },
    "workflow/jev_bench_tr/helpsteer2_verbosity": {
      "n": 1000,
      "accuracy": 0.211,
      "brier": 0.8591,
      "nll": 1.6434,
      "ece": 0.2384
    },
    "workflow/jev_bench_tr/ledgar": {
      "n": 1000,
      "accuracy": 0.087,
      "brier": 1.1279,
      "nll": 5.1482,
      "ece": 0.2662
    },
    "workflow/jev_bench_tr/massive": {
      "n": 1000,
      "accuracy": 0.376,
      "brier": 0.841,
      "nll": 3.0543,
      "ece": 0.2749
    },
    "workflow/jev_bench_tr/measuring_hate_speech": {
      "n": 1000,
      "accuracy": 0.315,
      "brier": 0.5364,
      "nll": 1.2327,
      "ece": 0.1636
    },
    "workflow/jev_bench_tr/mmlu": {
      "n": 1000,
      "accuracy": 0.274,
      "brier": 0.8533,
      "nll": 1.632,
      "ece": 0.2128
    },
    "workflow/jev_bench_tr/mnli": {
      "n": 1000,
      "accuracy": 0.719,
      "brier": 0.4591,
      "nll": 1.0263,
      "ece": 0.1899
    },
    "workflow/jev_bench_tr/paws": {
      "n": 1000,
      "accuracy": 0.526,
      "brier": 0.8253,
      "nll": 1.7845,
      "ece": 0.3995
    },
    "workflow/jev_bench_tr/sms_spam": {
      "n": 800,
      "accuracy": 0.3262,
      "brier": 1.2111,
      "nll": 4.8523,
      "ece": 0.6033
    },
    "workflow/jev_bench_tr/sst5": {
      "n": 1000,
      "accuracy": 0.283,
      "brier": 0.819,
      "nll": 1.6126,
      "ece": 0.1645
    },
    "workflow/jev_bench_tr/strategyqa_closed": {
      "n": 687,
      "accuracy": 0.5167,
      "brier": 0.8118,
      "nll": 1.415,
      "ece": 0.378
    },
    "workflow/jev_bench_tr/strategyqa_grounded": {
      "n": 687,
      "accuracy": 0.5459,
      "brier": 0.7267,
      "nll": 1.3457,
      "ece": 0.3306
    },
    "workflow/jev_bench_tr/stsb": {
      "n": 1000,
      "accuracy": 0.157,
      "brier": 1.159,
      "nll": 2.5626,
      "ece": 0.4587
    },
    "workflow/jev_bench_tr/yelp5": {
      "n": 1000,
      "accuracy": 0.189,
      "brier": 1.1719,
      "nll": 3.2098,
      "ece": 0.4934
    },
    "workflow/jev_bench_tr_en/arc_challenge": {
      "n": 91,
      "accuracy": 0.3077,
      "brier": 0.8524,
      "nll": 1.6184,
      "ece": 0.2614
    },
    "workflow/jev_bench_tr_en/banking77": {
      "n": 71,
      "accuracy": 0.4366,
      "brier": 0.9079,
      "nll": 4.9125,
      "ece": 0.4129
    },
    "workflow/jev_bench_tr_en/boolq": {
      "n": 80,
      "accuracy": 0.725,
      "brier": 0.4525,
      "nll": 0.8849,
      "ece": 0.2192
    },
    "workflow/jev_bench_tr_en/chaosnli": {
      "n": 151,
      "accuracy": 0.6159,
      "brier": 0.3676,
      "nll": 1.9762,
      "ece": 0.2812
    },
    "workflow/jev_bench_tr_en/civil_comments": {
      "n": 171,
      "accuracy": 0.9532,
      "brier": 0.0364,
      "nll": 0.253,
      "ece": 0.0843
    },
    "workflow/jev_bench_tr_en/clinc150": {
      "n": 96,
      "accuracy": 0.4792,
      "brier": 0.7581,
      "nll": 2.9066,
      "ece": 0.309
    },
    "workflow/jev_bench_tr_en/fever_evidence": {
      "n": 91,
      "accuracy": 0.7253,
      "brier": 0.4522,
      "nll": 0.7519,
      "ece": 0.2242
    },
    "workflow/jev_bench_tr_en/go_emotions": {
      "n": 93,
      "accuracy": 0.3978,
      "brier": 0.5753,
      "nll": 4.0598,
      "ece": 0.3649
    },
    "workflow/jev_bench_tr_en/helpsteer2_helpfulness": {
      "n": 89,
      "accuracy": 0.1348,
      "brier": 0.8934,
      "nll": 1.789,
      "ece": 0.314
    },
    "workflow/jev_bench_tr_en/helpsteer2_verbosity": {
      "n": 79,
      "accuracy": 0.1646,
      "brier": 0.9506,
      "nll": 1.6858,
      "ece": 0.3924
    },
    "workflow/jev_bench_tr_en/ledgar": {
      "n": 70,
      "accuracy": 0.2714,
      "brier": 0.9941,
      "nll": 4.4465,
      "ece": 0.3259
    },
    "workflow/jev_bench_tr_en/massive": {
      "n": 95,
      "accuracy": 0.4105,
      "brier": 0.8462,
      "nll": 3.0372,
      "ece": 0.2842
    },
    "workflow/jev_bench_tr_en/measuring_hate_speech": {
      "n": 96,
      "accuracy": 0.4479,
      "brier": 0.4822,
      "nll": 1.085,
      "ece": 0.0743
    },
    "workflow/jev_bench_tr_en/mmlu": {
      "n": 80,
      "accuracy": 0.35,
      "brier": 0.789,
      "nll": 1.4979,
      "ece": 0.1662
    },
    "workflow/jev_bench_tr_en/mnli": {
      "n": 92,
      "accuracy": 0.8913,
      "brier": 0.1695,
      "nll": 0.3677,
      "ece": 0.0731
    },
    "workflow/jev_bench_tr_en/paws": {
      "n": 80,
      "accuracy": 0.65,
      "brier": 0.4396,
      "nll": 0.679,
      "ece": 0.2589
    },
    "workflow/jev_bench_tr_en/sms_spam": {
      "n": 82,
      "accuracy": 0.5488,
      "brier": 0.7825,
      "nll": 2.6591,
      "ece": 0.3926
    },
    "workflow/jev_bench_tr_en/sst5": {
      "n": 73,
      "accuracy": 0.3014,
      "brier": 1.0188,
      "nll": 2.3765,
      "ece": 0.4116
    },
    "workflow/jev_bench_tr_en/strategyqa_closed": {
      "n": 60,
      "accuracy": 0.5333,
      "brier": 0.5837,
      "nll": 0.8485,
      "ece": 0.2996
    },
    "workflow/jev_bench_tr_en/strategyqa_grounded": {
      "n": 71,
      "accuracy": 0.5915,
      "brier": 0.5671,
      "nll": 0.936,
      "ece": 0.2506
    },
    "workflow/jev_bench_tr_en/stsb": {
      "n": 90,
      "accuracy": 0.2333,
      "brier": 1.0747,
      "nll": 2.5196,
      "ece": 0.3883
    },
    "workflow/jev_bench_tr_en/yelp5": {
      "n": 99,
      "accuracy": 0.2727,
      "brier": 1.1901,
      "nll": 4.5967,
      "ece": 0.5004
    },
    "workflow/massive_tr": {
      "n": 2974,
      "accuracy": 0.3534,
      "brier": 0.8701,
      "nll": 3.228,
      "ece": 0.2862
    },
    "workflow/squad_tr": {
      "n": 11873,
      "accuracy": 0.5198,
      "brier": 0.7361,
      "nll": 1.2571,
      "ece": 0.319
    },
    "workflow/xnli_tr": {
      "n": 5010,
      "accuracy": 0.6912,
      "brier": 0.5108,
      "nll": 1.1705,
      "ece": 0.2123
    }
  }
}