{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.995008319467554,
  "eval_steps": 500,
  "global_step": 18000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008319467554076539,
      "grad_norm": 0.5583499073982239,
      "learning_rate": 0.0004999994583628932,
      "loss": 0.7169,
      "step": 50
    },
    {
      "epoch": 0.016638935108153077,
      "grad_norm": 0.19315695762634277,
      "learning_rate": 0.0004999989167257863,
      "loss": 0.6946,
      "step": 100
    },
    {
      "epoch": 0.024958402662229616,
      "grad_norm": 0.2500464618206024,
      "learning_rate": 0.0004999983750886796,
      "loss": 0.6934,
      "step": 150
    },
    {
      "epoch": 0.033277870216306155,
      "grad_norm": 0.10472989827394485,
      "learning_rate": 0.0004999978334515727,
      "loss": 0.6922,
      "step": 200
    },
    {
      "epoch": 0.04159733777038269,
      "grad_norm": 0.08617454022169113,
      "learning_rate": 0.0004999972918144659,
      "loss": 0.6891,
      "step": 250
    },
    {
      "epoch": 0.04991680532445923,
      "grad_norm": 0.25109192728996277,
      "learning_rate": 0.0004999967501773591,
      "loss": 0.687,
      "step": 300
    },
    {
      "epoch": 0.05823627287853577,
      "grad_norm": 0.06475350260734558,
      "learning_rate": 0.0004999962085402523,
      "loss": 0.6854,
      "step": 350
    },
    {
      "epoch": 0.06655574043261231,
      "grad_norm": 0.5543227791786194,
      "learning_rate": 0.0004999956669031455,
      "loss": 0.6836,
      "step": 400
    },
    {
      "epoch": 0.07487520798668885,
      "grad_norm": 0.39940813183784485,
      "learning_rate": 0.0004999951252660387,
      "loss": 0.6808,
      "step": 450
    },
    {
      "epoch": 0.08319467554076539,
      "grad_norm": 0.3018794655799866,
      "learning_rate": 0.0004999945836289318,
      "loss": 0.6762,
      "step": 500
    },
    {
      "epoch": 0.08319467554076539,
      "eval_accuracy": 0.5590276660651246,
      "eval_auc": 0.5967892656129481,
      "eval_f1": 0.4315189357947604,
      "eval_loss": 0.6786146759986877,
      "eval_precision": 0.6065314861727951,
      "eval_recall": 0.3348881306732058,
      "eval_runtime": 3130.7158,
      "eval_samples_per_second": 109.208,
      "eval_steps_per_second": 13.651,
      "step": 500
    },
    {
      "epoch": 0.09151414309484193,
      "grad_norm": 0.2409348487854004,
      "learning_rate": 0.000499994041991825,
      "loss": 0.6738,
      "step": 550
    },
    {
      "epoch": 0.09983361064891846,
      "grad_norm": 0.292751282453537,
      "learning_rate": 0.0004999935003547182,
      "loss": 0.6718,
      "step": 600
    },
    {
      "epoch": 0.10815307820299501,
      "grad_norm": 0.4114605486392975,
      "learning_rate": 0.0004999929587176114,
      "loss": 0.6691,
      "step": 650
    },
    {
      "epoch": 0.11647254575707154,
      "grad_norm": 0.30100876092910767,
      "learning_rate": 0.0004999924170805046,
      "loss": 0.6691,
      "step": 700
    },
    {
      "epoch": 0.12479201331114809,
      "grad_norm": 0.20736241340637207,
      "learning_rate": 0.0004999918754433977,
      "loss": 0.6653,
      "step": 750
    },
    {
      "epoch": 0.13311148086522462,
      "grad_norm": 0.11416225880384445,
      "learning_rate": 0.0004999913338062909,
      "loss": 0.6629,
      "step": 800
    },
    {
      "epoch": 0.14143094841930118,
      "grad_norm": 0.23378710448741913,
      "learning_rate": 0.000499990792169184,
      "loss": 0.66,
      "step": 850
    },
    {
      "epoch": 0.1497504159733777,
      "grad_norm": 0.3658539652824402,
      "learning_rate": 0.0004999902505320771,
      "loss": 0.6538,
      "step": 900
    },
    {
      "epoch": 0.15806988352745424,
      "grad_norm": 0.23335859179496765,
      "learning_rate": 0.0004999897088949704,
      "loss": 0.6572,
      "step": 950
    },
    {
      "epoch": 0.16638935108153077,
      "grad_norm": 0.11527813971042633,
      "learning_rate": 0.0004999891672578635,
      "loss": 0.6559,
      "step": 1000
    },
    {
      "epoch": 0.16638935108153077,
      "eval_accuracy": 0.6038917926054186,
      "eval_auc": 0.6553736533116105,
      "eval_f1": 0.5857262157127257,
      "eval_loss": 0.650829017162323,
      "eval_precision": 0.6135620810315436,
      "eval_recall": 0.5603064335836225,
      "eval_runtime": 3082.7349,
      "eval_samples_per_second": 110.908,
      "eval_steps_per_second": 13.864,
      "step": 1000
    },
    {
      "epoch": 0.17470881863560733,
      "grad_norm": 0.12938088178634644,
      "learning_rate": 0.0004999886256207567,
      "loss": 0.6453,
      "step": 1050
    },
    {
      "epoch": 0.18302828618968386,
      "grad_norm": 0.16796837747097015,
      "learning_rate": 0.0004999880839836499,
      "loss": 0.6459,
      "step": 1100
    },
    {
      "epoch": 0.1913477537437604,
      "grad_norm": 0.31427496671676636,
      "learning_rate": 0.0004999875423465431,
      "loss": 0.6404,
      "step": 1150
    },
    {
      "epoch": 0.19966722129783693,
      "grad_norm": 0.24732699990272522,
      "learning_rate": 0.0004999870007094363,
      "loss": 0.638,
      "step": 1200
    },
    {
      "epoch": 0.2079866888519135,
      "grad_norm": 0.19166573882102966,
      "learning_rate": 0.0004999864590723295,
      "loss": 0.6316,
      "step": 1250
    },
    {
      "epoch": 0.21630615640599002,
      "grad_norm": 0.23063616454601288,
      "learning_rate": 0.0004999859174352226,
      "loss": 0.6243,
      "step": 1300
    },
    {
      "epoch": 0.22462562396006655,
      "grad_norm": 0.21349310874938965,
      "learning_rate": 0.0004999853757981158,
      "loss": 0.6229,
      "step": 1350
    },
    {
      "epoch": 0.23294509151414308,
      "grad_norm": 0.21778538823127747,
      "learning_rate": 0.000499984834161009,
      "loss": 0.6173,
      "step": 1400
    },
    {
      "epoch": 0.24126455906821964,
      "grad_norm": 0.3121509850025177,
      "learning_rate": 0.0004999842925239022,
      "loss": 0.6061,
      "step": 1450
    },
    {
      "epoch": 0.24958402662229617,
      "grad_norm": 0.30077889561653137,
      "learning_rate": 0.0004999837508867954,
      "loss": 0.597,
      "step": 1500
    },
    {
      "epoch": 0.24958402662229617,
      "eval_accuracy": 0.6591010795585831,
      "eval_auc": 0.7267389932367714,
      "eval_f1": 0.6891016322077618,
      "eval_loss": 0.5957785844802856,
      "eval_precision": 0.6331130956465479,
      "eval_recall": 0.7559533911944238,
      "eval_runtime": 3116.4637,
      "eval_samples_per_second": 109.707,
      "eval_steps_per_second": 13.714,
      "step": 1500
    },
    {
      "epoch": 0.2579034941763727,
      "grad_norm": 0.29947781562805176,
      "learning_rate": 0.0004999832092496885,
      "loss": 0.594,
      "step": 1550
    },
    {
      "epoch": 0.26622296173044924,
      "grad_norm": 0.35637083649635315,
      "learning_rate": 0.0004999826676125817,
      "loss": 0.5865,
      "step": 1600
    },
    {
      "epoch": 0.27454242928452577,
      "grad_norm": 0.4300802946090698,
      "learning_rate": 0.0004999821259754749,
      "loss": 0.5795,
      "step": 1650
    },
    {
      "epoch": 0.28286189683860236,
      "grad_norm": 0.37862110137939453,
      "learning_rate": 0.0004999815843383681,
      "loss": 0.5767,
      "step": 1700
    },
    {
      "epoch": 0.2911813643926789,
      "grad_norm": 0.5083811283111572,
      "learning_rate": 0.0004999810427012613,
      "loss": 0.5665,
      "step": 1750
    },
    {
      "epoch": 0.2995008319467554,
      "grad_norm": 0.4686565399169922,
      "learning_rate": 0.0004999805010641545,
      "loss": 0.5628,
      "step": 1800
    },
    {
      "epoch": 0.30782029950083195,
      "grad_norm": 0.6207341551780701,
      "learning_rate": 0.0004999799594270476,
      "loss": 0.5612,
      "step": 1850
    },
    {
      "epoch": 0.3161397670549085,
      "grad_norm": 0.400837779045105,
      "learning_rate": 0.0004999794177899408,
      "loss": 0.5546,
      "step": 1900
    },
    {
      "epoch": 0.324459234608985,
      "grad_norm": 0.3956456184387207,
      "learning_rate": 0.000499978876152834,
      "loss": 0.5522,
      "step": 1950
    },
    {
      "epoch": 0.33277870216306155,
      "grad_norm": 0.39825043082237244,
      "learning_rate": 0.0004999783345157272,
      "loss": 0.547,
      "step": 2000
    },
    {
      "epoch": 0.33277870216306155,
      "eval_accuracy": 0.696047663198781,
      "eval_auc": 0.7729761853730717,
      "eval_f1": 0.7419938676961655,
      "eval_loss": 0.5397406220436096,
      "eval_precision": 0.644336742613705,
      "eval_recall": 0.8745413152766154,
      "eval_runtime": 3213.3641,
      "eval_samples_per_second": 106.399,
      "eval_steps_per_second": 13.3,
      "step": 2000
    },
    {
      "epoch": 0.3410981697171381,
      "grad_norm": 0.4658432900905609,
      "learning_rate": 0.0004999777928786204,
      "loss": 0.5417,
      "step": 2050
    },
    {
      "epoch": 0.34941763727121466,
      "grad_norm": 0.3485628664493561,
      "learning_rate": 0.0004999772512415135,
      "loss": 0.5391,
      "step": 2100
    },
    {
      "epoch": 0.3577371048252912,
      "grad_norm": 0.3111945390701294,
      "learning_rate": 0.0004999767096044067,
      "loss": 0.5395,
      "step": 2150
    },
    {
      "epoch": 0.36605657237936773,
      "grad_norm": 0.33330038189888,
      "learning_rate": 0.0004999761679672999,
      "loss": 0.5341,
      "step": 2200
    },
    {
      "epoch": 0.37437603993344426,
      "grad_norm": 0.6060266494750977,
      "learning_rate": 0.0004999756263301931,
      "loss": 0.5348,
      "step": 2250
    },
    {
      "epoch": 0.3826955074875208,
      "grad_norm": 0.38933542370796204,
      "learning_rate": 0.0004999750846930863,
      "loss": 0.5346,
      "step": 2300
    },
    {
      "epoch": 0.3910149750415973,
      "grad_norm": 0.30752280354499817,
      "learning_rate": 0.0004999745430559793,
      "loss": 0.5292,
      "step": 2350
    },
    {
      "epoch": 0.39933444259567386,
      "grad_norm": 0.3584199845790863,
      "learning_rate": 0.0004999740014188725,
      "loss": 0.528,
      "step": 2400
    },
    {
      "epoch": 0.40765391014975044,
      "grad_norm": 0.34107691049575806,
      "learning_rate": 0.0004999734597817657,
      "loss": 0.5255,
      "step": 2450
    },
    {
      "epoch": 0.415973377703827,
      "grad_norm": 0.33743101358413696,
      "learning_rate": 0.0004999729181446589,
      "loss": 0.5241,
      "step": 2500
    },
    {
      "epoch": 0.415973377703827,
      "eval_accuracy": 0.7029268877651003,
      "eval_auc": 0.7850390034483034,
      "eval_f1": 0.7599834583801028,
      "eval_loss": 0.5229647755622864,
      "eval_precision": 0.6373332382107441,
      "eval_recall": 0.9410893725602655,
      "eval_runtime": 3103.5803,
      "eval_samples_per_second": 110.163,
      "eval_steps_per_second": 13.771,
      "step": 2500
    },
    {
      "epoch": 0.4242928452579035,
      "grad_norm": 0.43357181549072266,
      "learning_rate": 0.0004999723765075521,
      "loss": 0.526,
      "step": 2550
    },
    {
      "epoch": 0.43261231281198004,
      "grad_norm": 0.536624550819397,
      "learning_rate": 0.0004999718348704453,
      "loss": 0.5235,
      "step": 2600
    },
    {
      "epoch": 0.44093178036605657,
      "grad_norm": 0.26722005009651184,
      "learning_rate": 0.0004999712932333384,
      "loss": 0.5238,
      "step": 2650
    },
    {
      "epoch": 0.4492512479201331,
      "grad_norm": 0.29940328001976013,
      "learning_rate": 0.0004999707515962316,
      "loss": 0.5127,
      "step": 2700
    },
    {
      "epoch": 0.45757071547420963,
      "grad_norm": 0.27166950702667236,
      "learning_rate": 0.0004999702099591248,
      "loss": 0.5167,
      "step": 2750
    },
    {
      "epoch": 0.46589018302828616,
      "grad_norm": 0.5428174138069153,
      "learning_rate": 0.000499969668322018,
      "loss": 0.5148,
      "step": 2800
    },
    {
      "epoch": 0.47420965058236275,
      "grad_norm": 0.37499845027923584,
      "learning_rate": 0.0004999691266849112,
      "loss": 0.5135,
      "step": 2850
    },
    {
      "epoch": 0.4825291181364393,
      "grad_norm": 0.5476682186126709,
      "learning_rate": 0.0004999685850478043,
      "loss": 0.5127,
      "step": 2900
    },
    {
      "epoch": 0.4908485856905158,
      "grad_norm": 0.24501897394657135,
      "learning_rate": 0.0004999680434106975,
      "loss": 0.5098,
      "step": 2950
    },
    {
      "epoch": 0.49916805324459235,
      "grad_norm": 0.3694550395011902,
      "learning_rate": 0.0004999675017735907,
      "loss": 0.5138,
      "step": 3000
    },
    {
      "epoch": 0.49916805324459235,
      "eval_accuracy": 0.7139476863050199,
      "eval_auc": 0.794041583180005,
      "eval_f1": 0.7427284708283901,
      "eval_loss": 0.5122376084327698,
      "eval_precision": 0.6745716224352297,
      "eval_recall": 0.8262060408851225,
      "eval_runtime": 3072.5721,
      "eval_samples_per_second": 111.275,
      "eval_steps_per_second": 13.91,
      "step": 3000
    },
    {
      "epoch": 0.5074875207986689,
      "grad_norm": 0.2761985957622528,
      "learning_rate": 0.0004999669601364839,
      "loss": 0.5187,
      "step": 3050
    },
    {
      "epoch": 0.5158069883527454,
      "grad_norm": 0.23015080392360687,
      "learning_rate": 0.0004999664184993771,
      "loss": 0.5117,
      "step": 3100
    },
    {
      "epoch": 0.5241264559068219,
      "grad_norm": 0.26510143280029297,
      "learning_rate": 0.0004999658768622703,
      "loss": 0.5124,
      "step": 3150
    },
    {
      "epoch": 0.5324459234608985,
      "grad_norm": 0.6037235260009766,
      "learning_rate": 0.0004999653352251634,
      "loss": 0.5111,
      "step": 3200
    },
    {
      "epoch": 0.540765391014975,
      "grad_norm": 0.4400153160095215,
      "learning_rate": 0.0004999647935880566,
      "loss": 0.5167,
      "step": 3250
    },
    {
      "epoch": 0.5490848585690515,
      "grad_norm": 0.33088386058807373,
      "learning_rate": 0.0004999642519509498,
      "loss": 0.5103,
      "step": 3300
    },
    {
      "epoch": 0.5574043261231281,
      "grad_norm": 0.45530447363853455,
      "learning_rate": 0.000499963710313843,
      "loss": 0.5136,
      "step": 3350
    },
    {
      "epoch": 0.5657237936772047,
      "grad_norm": 0.38819849491119385,
      "learning_rate": 0.0004999631686767362,
      "loss": 0.5094,
      "step": 3400
    },
    {
      "epoch": 0.5740432612312812,
      "grad_norm": 0.2275410294532776,
      "learning_rate": 0.0004999626270396293,
      "loss": 0.5095,
      "step": 3450
    },
    {
      "epoch": 0.5823627287853578,
      "grad_norm": 0.25340116024017334,
      "learning_rate": 0.0004999620854025226,
      "loss": 0.5109,
      "step": 3500
    },
    {
      "epoch": 0.5823627287853578,
      "eval_accuracy": 0.7234504926893615,
      "eval_auc": 0.8032986510196456,
      "eval_f1": 0.7607974094312893,
      "eval_loss": 0.5012865662574768,
      "eval_precision": 0.6700384562254078,
      "eval_recall": 0.8799957862456034,
      "eval_runtime": 3072.7802,
      "eval_samples_per_second": 111.267,
      "eval_steps_per_second": 13.909,
      "step": 3500
    },
    {
      "epoch": 0.5906821963394343,
      "grad_norm": 0.2851439416408539,
      "learning_rate": 0.0004999615437654157,
      "loss": 0.5072,
      "step": 3550
    },
    {
      "epoch": 0.5990016638935108,
      "grad_norm": 0.2701944410800934,
      "learning_rate": 0.0004999610021283089,
      "loss": 0.5013,
      "step": 3600
    },
    {
      "epoch": 0.6073211314475874,
      "grad_norm": 0.3663853704929352,
      "learning_rate": 0.0004999604604912021,
      "loss": 0.509,
      "step": 3650
    },
    {
      "epoch": 0.6156405990016639,
      "grad_norm": 0.3946699798107147,
      "learning_rate": 0.0004999599188540951,
      "loss": 0.5073,
      "step": 3700
    },
    {
      "epoch": 0.6239600665557404,
      "grad_norm": 0.20986106991767883,
      "learning_rate": 0.0004999593772169885,
      "loss": 0.5032,
      "step": 3750
    },
    {
      "epoch": 0.632279534109817,
      "grad_norm": 0.44309139251708984,
      "learning_rate": 0.0004999588355798815,
      "loss": 0.5062,
      "step": 3800
    },
    {
      "epoch": 0.6405990016638935,
      "grad_norm": 0.3160322904586792,
      "learning_rate": 0.0004999582939427747,
      "loss": 0.5003,
      "step": 3850
    },
    {
      "epoch": 0.64891846921797,
      "grad_norm": 0.36600473523139954,
      "learning_rate": 0.0004999577523056679,
      "loss": 0.5047,
      "step": 3900
    },
    {
      "epoch": 0.6572379367720466,
      "grad_norm": 0.4690772593021393,
      "learning_rate": 0.0004999572106685611,
      "loss": 0.504,
      "step": 3950
    },
    {
      "epoch": 0.6655574043261231,
      "grad_norm": 0.2607940137386322,
      "learning_rate": 0.0004999566690314542,
      "loss": 0.5039,
      "step": 4000
    },
    {
      "epoch": 0.6655574043261231,
      "eval_accuracy": 0.7287415289310586,
      "eval_auc": 0.8093233697055171,
      "eval_f1": 0.7662815165757343,
      "eval_loss": 0.49741330742836,
      "eval_precision": 0.6728864418931957,
      "eval_recall": 0.8897810603444745,
      "eval_runtime": 3108.812,
      "eval_samples_per_second": 109.977,
      "eval_steps_per_second": 13.747,
      "step": 4000
    },
    {
      "epoch": 0.6738768718801996,
      "grad_norm": 0.4410100281238556,
      "learning_rate": 0.0004999561273943475,
      "loss": 0.5004,
      "step": 4050
    },
    {
      "epoch": 0.6821963394342762,
      "grad_norm": 0.3221227824687958,
      "learning_rate": 0.0004999555857572406,
      "loss": 0.4987,
      "step": 4100
    },
    {
      "epoch": 0.6905158069883528,
      "grad_norm": 0.3825673460960388,
      "learning_rate": 0.0004999550441201338,
      "loss": 0.5006,
      "step": 4150
    },
    {
      "epoch": 0.6988352745424293,
      "grad_norm": 0.4332997798919678,
      "learning_rate": 0.000499954502483027,
      "loss": 0.498,
      "step": 4200
    },
    {
      "epoch": 0.7071547420965059,
      "grad_norm": 0.4402281939983368,
      "learning_rate": 0.0004999539608459201,
      "loss": 0.497,
      "step": 4250
    },
    {
      "epoch": 0.7154742096505824,
      "grad_norm": 0.2534608840942383,
      "learning_rate": 0.0004999534192088134,
      "loss": 0.4959,
      "step": 4300
    },
    {
      "epoch": 0.7237936772046589,
      "grad_norm": 0.43906906247138977,
      "learning_rate": 0.0004999528775717065,
      "loss": 0.4923,
      "step": 4350
    },
    {
      "epoch": 0.7321131447587355,
      "grad_norm": 0.3567379117012024,
      "learning_rate": 0.0004999523359345997,
      "loss": 0.497,
      "step": 4400
    },
    {
      "epoch": 0.740432612312812,
      "grad_norm": 0.26709049940109253,
      "learning_rate": 0.0004999517942974929,
      "loss": 0.494,
      "step": 4450
    },
    {
      "epoch": 0.7487520798668885,
      "grad_norm": 0.3749837577342987,
      "learning_rate": 0.0004999512526603861,
      "loss": 0.4954,
      "step": 4500
    },
    {
      "epoch": 0.7487520798668885,
      "eval_accuracy": 0.7357143483894366,
      "eval_auc": 0.8154262901262641,
      "eval_f1": 0.7617121262865868,
      "eval_loss": 0.48994818329811096,
      "eval_precision": 0.6932252366415145,
      "eval_recall": 0.8452147551633122,
      "eval_runtime": 3064.0111,
      "eval_samples_per_second": 111.585,
      "eval_steps_per_second": 13.948,
      "step": 4500
    },
    {
      "epoch": 0.757071547420965,
      "grad_norm": 0.34779804944992065,
      "learning_rate": 0.0004999507110232793,
      "loss": 0.4947,
      "step": 4550
    },
    {
      "epoch": 0.7653910149750416,
      "grad_norm": 0.48780205845832825,
      "learning_rate": 0.0004999501693861724,
      "loss": 0.5003,
      "step": 4600
    },
    {
      "epoch": 0.7737104825291181,
      "grad_norm": 0.25045379996299744,
      "learning_rate": 0.0004999496277490656,
      "loss": 0.4922,
      "step": 4650
    },
    {
      "epoch": 0.7820299500831946,
      "grad_norm": 0.29499462246894836,
      "learning_rate": 0.0004999490861119588,
      "loss": 0.4891,
      "step": 4700
    },
    {
      "epoch": 0.7903494176372712,
      "grad_norm": 0.34115445613861084,
      "learning_rate": 0.000499948544474852,
      "loss": 0.4984,
      "step": 4750
    },
    {
      "epoch": 0.7986688851913477,
      "grad_norm": 0.5018367171287537,
      "learning_rate": 0.0004999480028377452,
      "loss": 0.4935,
      "step": 4800
    },
    {
      "epoch": 0.8069883527454242,
      "grad_norm": 0.492158442735672,
      "learning_rate": 0.0004999474612006384,
      "loss": 0.4914,
      "step": 4850
    },
    {
      "epoch": 0.8153078202995009,
      "grad_norm": 0.3154297173023224,
      "learning_rate": 0.0004999469195635315,
      "loss": 0.4946,
      "step": 4900
    },
    {
      "epoch": 0.8236272878535774,
      "grad_norm": 0.27449771761894226,
      "learning_rate": 0.0004999463779264247,
      "loss": 0.4906,
      "step": 4950
    },
    {
      "epoch": 0.831946755407654,
      "grad_norm": 0.37241068482398987,
      "learning_rate": 0.0004999458362893179,
      "loss": 0.4906,
      "step": 5000
    },
    {
      "epoch": 0.831946755407654,
      "eval_accuracy": 0.7396014612502523,
      "eval_auc": 0.8197279172699253,
      "eval_f1": 0.7709238179535209,
      "eval_loss": 0.48505428433418274,
      "eval_precision": 0.6878946087547468,
      "eval_recall": 0.876747683898191,
      "eval_runtime": 3075.4729,
      "eval_samples_per_second": 111.17,
      "eval_steps_per_second": 13.896,
      "step": 5000
    },
    {
      "epoch": 0.8402662229617305,
      "grad_norm": 0.2525058090686798,
      "learning_rate": 0.0004999452946522111,
      "loss": 0.4907,
      "step": 5050
    },
    {
      "epoch": 0.848585690515807,
      "grad_norm": 0.3323892056941986,
      "learning_rate": 0.0004999447530151043,
      "loss": 0.4914,
      "step": 5100
    },
    {
      "epoch": 0.8569051580698835,
      "grad_norm": 0.28021496534347534,
      "learning_rate": 0.0004999442113779973,
      "loss": 0.4891,
      "step": 5150
    },
    {
      "epoch": 0.8652246256239601,
      "grad_norm": 0.2550656497478485,
      "learning_rate": 0.0004999436697408905,
      "loss": 0.4866,
      "step": 5200
    },
    {
      "epoch": 0.8735440931780366,
      "grad_norm": 0.27683132886886597,
      "learning_rate": 0.0004999431281037837,
      "loss": 0.4867,
      "step": 5250
    },
    {
      "epoch": 0.8818635607321131,
      "grad_norm": 0.3277775049209595,
      "learning_rate": 0.0004999425864666769,
      "loss": 0.4897,
      "step": 5300
    },
    {
      "epoch": 0.8901830282861897,
      "grad_norm": 0.29307055473327637,
      "learning_rate": 0.0004999420448295701,
      "loss": 0.4861,
      "step": 5350
    },
    {
      "epoch": 0.8985024958402662,
      "grad_norm": 0.4498750865459442,
      "learning_rate": 0.0004999415031924633,
      "loss": 0.4839,
      "step": 5400
    },
    {
      "epoch": 0.9068219633943427,
      "grad_norm": 0.2954080104827881,
      "learning_rate": 0.0004999409615553564,
      "loss": 0.4857,
      "step": 5450
    },
    {
      "epoch": 0.9151414309484193,
      "grad_norm": 0.35260140895843506,
      "learning_rate": 0.0004999404199182496,
      "loss": 0.487,
      "step": 5500
    },
    {
      "epoch": 0.9151414309484193,
      "eval_accuracy": 0.7438073817121431,
      "eval_auc": 0.8241829377605762,
      "eval_f1": 0.7730318561781915,
      "eval_loss": 0.4809417724609375,
      "eval_precision": 0.6936197123499789,
      "eval_recall": 0.8729787146878603,
      "eval_runtime": 3075.8035,
      "eval_samples_per_second": 111.158,
      "eval_steps_per_second": 13.895,
      "step": 5500
    },
    {
      "epoch": 0.9234608985024958,
      "grad_norm": 0.353834867477417,
      "learning_rate": 0.0004999398782811428,
      "loss": 0.4848,
      "step": 5550
    },
    {
      "epoch": 0.9317803660565723,
      "grad_norm": 0.47889459133148193,
      "learning_rate": 0.000499939336644036,
      "loss": 0.4862,
      "step": 5600
    },
    {
      "epoch": 0.940099833610649,
      "grad_norm": 0.24000470340251923,
      "learning_rate": 0.0004999387950069292,
      "loss": 0.4807,
      "step": 5650
    },
    {
      "epoch": 0.9484193011647255,
      "grad_norm": 0.34989428520202637,
      "learning_rate": 0.0004999382533698223,
      "loss": 0.4862,
      "step": 5700
    },
    {
      "epoch": 0.956738768718802,
      "grad_norm": 0.3731101155281067,
      "learning_rate": 0.0004999377117327155,
      "loss": 0.4875,
      "step": 5750
    },
    {
      "epoch": 0.9650582362728786,
      "grad_norm": 0.26715531945228577,
      "learning_rate": 0.0004999371700956087,
      "loss": 0.4824,
      "step": 5800
    },
    {
      "epoch": 0.9733777038269551,
      "grad_norm": 0.25930115580558777,
      "learning_rate": 0.0004999366284585019,
      "loss": 0.4805,
      "step": 5850
    },
    {
      "epoch": 0.9816971713810316,
      "grad_norm": 0.3994530439376831,
      "learning_rate": 0.0004999360868213951,
      "loss": 0.4848,
      "step": 5900
    },
    {
      "epoch": 0.9900166389351082,
      "grad_norm": 0.3294505178928375,
      "learning_rate": 0.0004999355451842882,
      "loss": 0.4848,
      "step": 5950
    },
    {
      "epoch": 0.9983361064891847,
      "grad_norm": 0.3712209165096283,
      "learning_rate": 0.0004999350035471814,
      "loss": 0.4819,
      "step": 6000
    },
    {
      "epoch": 0.9983361064891847,
      "eval_accuracy": 0.7473610627699994,
      "eval_auc": 0.827144023087403,
      "eval_f1": 0.7763430113178821,
      "eval_loss": 0.4787616729736328,
      "eval_precision": 0.6961882470952102,
      "eval_recall": 0.8773563373110395,
      "eval_runtime": 3107.0522,
      "eval_samples_per_second": 110.04,
      "eval_steps_per_second": 13.755,
      "step": 6000
    },
    {
      "epoch": 1.0066555740432612,
      "grad_norm": 0.4111676812171936,
      "learning_rate": 0.0004999344619100746,
      "loss": 0.4793,
      "step": 6050
    },
    {
      "epoch": 1.0149750415973378,
      "grad_norm": 0.28988873958587646,
      "learning_rate": 0.0004999339202729678,
      "loss": 0.4809,
      "step": 6100
    },
    {
      "epoch": 1.0232945091514143,
      "grad_norm": 0.4895919859409332,
      "learning_rate": 0.000499933378635861,
      "loss": 0.4774,
      "step": 6150
    },
    {
      "epoch": 1.0316139767054908,
      "grad_norm": 0.40186378359794617,
      "learning_rate": 0.0004999328369987542,
      "loss": 0.4783,
      "step": 6200
    },
    {
      "epoch": 1.0399334442595674,
      "grad_norm": 0.31091171503067017,
      "learning_rate": 0.0004999322953616473,
      "loss": 0.472,
      "step": 6250
    },
    {
      "epoch": 1.0482529118136439,
      "grad_norm": 0.4297674000263214,
      "learning_rate": 0.0004999317645572826,
      "loss": 0.4755,
      "step": 6300
    },
    {
      "epoch": 1.0565723793677204,
      "grad_norm": 0.5054673552513123,
      "learning_rate": 0.0004999312229201758,
      "loss": 0.4715,
      "step": 6350
    },
    {
      "epoch": 1.064891846921797,
      "grad_norm": 0.3952338993549347,
      "learning_rate": 0.000499930681283069,
      "loss": 0.4742,
      "step": 6400
    },
    {
      "epoch": 1.0732113144758735,
      "grad_norm": 0.2842634320259094,
      "learning_rate": 0.0004999301396459622,
      "loss": 0.4772,
      "step": 6450
    },
    {
      "epoch": 1.08153078202995,
      "grad_norm": 0.3358585238456726,
      "learning_rate": 0.0004999295980088553,
      "loss": 0.4709,
      "step": 6500
    },
    {
      "epoch": 1.08153078202995,
      "eval_accuracy": 0.7469252615538508,
      "eval_auc": 0.8303769212813772,
      "eval_f1": 0.759462915601023,
      "eval_loss": 0.47938379645347595,
      "eval_precision": 0.7233056748441893,
      "eval_recall": 0.7994252907197912,
      "eval_runtime": 3090.922,
      "eval_samples_per_second": 110.614,
      "eval_steps_per_second": 13.827,
      "step": 6500
    },
    {
      "epoch": 1.0898502495840265,
      "grad_norm": 0.33011338114738464,
      "learning_rate": 0.0004999290563717485,
      "loss": 0.4741,
      "step": 6550
    },
    {
      "epoch": 1.098169717138103,
      "grad_norm": 0.3731289207935333,
      "learning_rate": 0.0004999285147346417,
      "loss": 0.4797,
      "step": 6600
    },
    {
      "epoch": 1.1064891846921796,
      "grad_norm": 0.3975057601928711,
      "learning_rate": 0.0004999279730975349,
      "loss": 0.4768,
      "step": 6650
    },
    {
      "epoch": 1.1148086522462561,
      "grad_norm": 0.37461864948272705,
      "learning_rate": 0.0004999274314604281,
      "loss": 0.4715,
      "step": 6700
    },
    {
      "epoch": 1.1231281198003327,
      "grad_norm": 0.4079597294330597,
      "learning_rate": 0.0004999268898233213,
      "loss": 0.4738,
      "step": 6750
    },
    {
      "epoch": 1.1314475873544092,
      "grad_norm": 0.25766322016716003,
      "learning_rate": 0.0004999263481862144,
      "loss": 0.4749,
      "step": 6800
    },
    {
      "epoch": 1.1397670549084857,
      "grad_norm": 0.32334184646606445,
      "learning_rate": 0.0004999258065491076,
      "loss": 0.4753,
      "step": 6850
    },
    {
      "epoch": 1.1480865224625625,
      "grad_norm": 0.3726251423358917,
      "learning_rate": 0.0004999252649120008,
      "loss": 0.4706,
      "step": 6900
    },
    {
      "epoch": 1.156405990016639,
      "grad_norm": 0.2608793079853058,
      "learning_rate": 0.000499924723274894,
      "loss": 0.4758,
      "step": 6950
    },
    {
      "epoch": 1.1647254575707155,
      "grad_norm": 0.2321689873933792,
      "learning_rate": 0.0004999241816377872,
      "loss": 0.4775,
      "step": 7000
    },
    {
      "epoch": 1.1647254575707155,
      "eval_accuracy": 0.7482034167985282,
      "eval_auc": 0.8299646263747171,
      "eval_f1": 0.7817299037815499,
      "eval_loss": 0.4775802195072174,
      "eval_precision": 0.6896254014833636,
      "eval_recall": 0.9022291931245574,
      "eval_runtime": 3102.1713,
      "eval_samples_per_second": 110.213,
      "eval_steps_per_second": 13.777,
      "step": 7000
    },
    {
      "epoch": 1.173044925124792,
      "grad_norm": 0.4716586172580719,
      "learning_rate": 0.0004999236400006803,
      "loss": 0.4688,
      "step": 7050
    },
    {
      "epoch": 1.1813643926788686,
      "grad_norm": 0.3553094267845154,
      "learning_rate": 0.0004999230983635735,
      "loss": 0.4699,
      "step": 7100
    },
    {
      "epoch": 1.1896838602329451,
      "grad_norm": 0.5420466065406799,
      "learning_rate": 0.0004999225567264667,
      "loss": 0.4742,
      "step": 7150
    },
    {
      "epoch": 1.1980033277870217,
      "grad_norm": 0.26752716302871704,
      "learning_rate": 0.0004999220150893599,
      "loss": 0.4725,
      "step": 7200
    },
    {
      "epoch": 1.2063227953410982,
      "grad_norm": 0.2967946529388428,
      "learning_rate": 0.0004999214734522531,
      "loss": 0.4671,
      "step": 7250
    },
    {
      "epoch": 1.2146422628951747,
      "grad_norm": 0.24301719665527344,
      "learning_rate": 0.0004999209318151461,
      "loss": 0.4741,
      "step": 7300
    },
    {
      "epoch": 1.2229617304492513,
      "grad_norm": 0.3372076451778412,
      "learning_rate": 0.0004999203901780393,
      "loss": 0.4718,
      "step": 7350
    },
    {
      "epoch": 1.2312811980033278,
      "grad_norm": 0.28806430101394653,
      "learning_rate": 0.0004999198485409325,
      "loss": 0.4751,
      "step": 7400
    },
    {
      "epoch": 1.2396006655574043,
      "grad_norm": 0.28929972648620605,
      "learning_rate": 0.0004999193069038257,
      "loss": 0.4748,
      "step": 7450
    },
    {
      "epoch": 1.2479201331114809,
      "grad_norm": 0.5165761113166809,
      "learning_rate": 0.0004999187652667189,
      "loss": 0.4738,
      "step": 7500
    },
    {
      "epoch": 1.2479201331114809,
      "eval_accuracy": 0.7527018212981027,
      "eval_auc": 0.8341791803614174,
      "eval_f1": 0.7692807265047057,
      "eval_loss": 0.4720609784126282,
      "eval_precision": 0.7206515404042986,
      "eval_recall": 0.8249477670027916,
      "eval_runtime": 3117.7578,
      "eval_samples_per_second": 109.662,
      "eval_steps_per_second": 13.708,
      "step": 7500
    },
    {
      "epoch": 1.2562396006655574,
      "grad_norm": 0.3871491551399231,
      "learning_rate": 0.0004999182236296121,
      "loss": 0.4749,
      "step": 7550
    },
    {
      "epoch": 1.264559068219634,
      "grad_norm": 0.34656795859336853,
      "learning_rate": 0.0004999176819925052,
      "loss": 0.4729,
      "step": 7600
    },
    {
      "epoch": 1.2728785357737105,
      "grad_norm": 0.4117598533630371,
      "learning_rate": 0.0004999171403553984,
      "loss": 0.4677,
      "step": 7650
    },
    {
      "epoch": 1.281198003327787,
      "grad_norm": 0.25794079899787903,
      "learning_rate": 0.0004999165987182916,
      "loss": 0.4723,
      "step": 7700
    },
    {
      "epoch": 1.2895174708818635,
      "grad_norm": 0.33428308367729187,
      "learning_rate": 0.0004999160570811848,
      "loss": 0.471,
      "step": 7750
    },
    {
      "epoch": 1.29783693843594,
      "grad_norm": 0.3722244203090668,
      "learning_rate": 0.000499915515444078,
      "loss": 0.4695,
      "step": 7800
    },
    {
      "epoch": 1.3061564059900166,
      "grad_norm": 0.34326881170272827,
      "learning_rate": 0.0004999149738069711,
      "loss": 0.4693,
      "step": 7850
    },
    {
      "epoch": 1.3144758735440931,
      "grad_norm": 0.473015695810318,
      "learning_rate": 0.0004999144321698643,
      "loss": 0.4773,
      "step": 7900
    },
    {
      "epoch": 1.3227953410981697,
      "grad_norm": 0.39536336064338684,
      "learning_rate": 0.0004999138905327575,
      "loss": 0.4633,
      "step": 7950
    },
    {
      "epoch": 1.3311148086522462,
      "grad_norm": 0.24282647669315338,
      "learning_rate": 0.0004999133488956507,
      "loss": 0.4704,
      "step": 8000
    },
    {
      "epoch": 1.3311148086522462,
      "eval_accuracy": 0.7551382133320074,
      "eval_auc": 0.8357908200441471,
      "eval_f1": 0.7845750840156658,
      "eval_loss": 0.46888816356658936,
      "eval_precision": 0.7001180258003481,
      "eval_recall": 0.8922039691225442,
      "eval_runtime": 3128.319,
      "eval_samples_per_second": 109.292,
      "eval_steps_per_second": 13.662,
      "step": 8000
    },
    {
      "epoch": 1.3394342762063227,
      "grad_norm": 0.46165645122528076,
      "learning_rate": 0.0004999128072585439,
      "loss": 0.4677,
      "step": 8050
    },
    {
      "epoch": 1.3477537437603995,
      "grad_norm": 0.2814394235610962,
      "learning_rate": 0.0004999122656214371,
      "loss": 0.4655,
      "step": 8100
    },
    {
      "epoch": 1.356073211314476,
      "grad_norm": 0.24799716472625732,
      "learning_rate": 0.0004999117239843302,
      "loss": 0.4705,
      "step": 8150
    },
    {
      "epoch": 1.3643926788685525,
      "grad_norm": 0.3806433379650116,
      "learning_rate": 0.0004999111823472234,
      "loss": 0.4689,
      "step": 8200
    },
    {
      "epoch": 1.372712146422629,
      "grad_norm": 0.3523135185241699,
      "learning_rate": 0.0004999106407101166,
      "loss": 0.4695,
      "step": 8250
    },
    {
      "epoch": 1.3810316139767056,
      "grad_norm": 0.2665747106075287,
      "learning_rate": 0.0004999100990730098,
      "loss": 0.4737,
      "step": 8300
    },
    {
      "epoch": 1.3893510815307821,
      "grad_norm": 0.28612571954727173,
      "learning_rate": 0.000499909557435903,
      "loss": 0.4686,
      "step": 8350
    },
    {
      "epoch": 1.3976705490848587,
      "grad_norm": 0.2960534691810608,
      "learning_rate": 0.0004999090157987961,
      "loss": 0.4696,
      "step": 8400
    },
    {
      "epoch": 1.4059900166389352,
      "grad_norm": 0.34513574838638306,
      "learning_rate": 0.0004999084741616894,
      "loss": 0.4694,
      "step": 8450
    },
    {
      "epoch": 1.4143094841930117,
      "grad_norm": 0.3929885923862457,
      "learning_rate": 0.0004999079325245825,
      "loss": 0.4687,
      "step": 8500
    },
    {
      "epoch": 1.4143094841930117,
      "eval_accuracy": 0.7571154054267488,
      "eval_auc": 0.8376226439602212,
      "eval_f1": 0.7821289144488288,
      "eval_loss": 0.4663051962852478,
      "eval_precision": 0.7088304808282172,
      "eval_recall": 0.8723349466550399,
      "eval_runtime": 3128.487,
      "eval_samples_per_second": 109.286,
      "eval_steps_per_second": 13.661,
      "step": 8500
    },
    {
      "epoch": 1.4226289517470883,
      "grad_norm": 0.667980968952179,
      "learning_rate": 0.0004999074017202178,
      "loss": 0.4712,
      "step": 8550
    },
    {
      "epoch": 1.4309484193011648,
      "grad_norm": 0.27172645926475525,
      "learning_rate": 0.000499906860083111,
      "loss": 0.4713,
      "step": 8600
    },
    {
      "epoch": 1.4392678868552413,
      "grad_norm": 0.4408021569252014,
      "learning_rate": 0.0004999063184460041,
      "loss": 0.4706,
      "step": 8650
    },
    {
      "epoch": 1.4475873544093179,
      "grad_norm": 0.25165098905563354,
      "learning_rate": 0.0004999057768088973,
      "loss": 0.4699,
      "step": 8700
    },
    {
      "epoch": 1.4559068219633944,
      "grad_norm": 0.35004204511642456,
      "learning_rate": 0.0004999052351717905,
      "loss": 0.4708,
      "step": 8750
    },
    {
      "epoch": 1.464226289517471,
      "grad_norm": 0.33756914734840393,
      "learning_rate": 0.0004999046935346837,
      "loss": 0.4665,
      "step": 8800
    },
    {
      "epoch": 1.4725457570715474,
      "grad_norm": 0.4379380941390991,
      "learning_rate": 0.0004999041518975769,
      "loss": 0.4671,
      "step": 8850
    },
    {
      "epoch": 1.480865224625624,
      "grad_norm": 0.450702965259552,
      "learning_rate": 0.0004999036102604701,
      "loss": 0.4682,
      "step": 8900
    },
    {
      "epoch": 1.4891846921797005,
      "grad_norm": 0.429463654756546,
      "learning_rate": 0.0004999030686233632,
      "loss": 0.4657,
      "step": 8950
    },
    {
      "epoch": 1.497504159733777,
      "grad_norm": 0.26801300048828125,
      "learning_rate": 0.0004999025269862564,
      "loss": 0.4663,
      "step": 9000
    },
    {
      "epoch": 1.497504159733777,
      "eval_accuracy": 0.7582911912582371,
      "eval_auc": 0.839204544655265,
      "eval_f1": 0.7839499717650011,
      "eval_loss": 0.4652154743671417,
      "eval_precision": 0.7084461454863326,
      "eval_recall": 0.8774675336076175,
      "eval_runtime": 3125.7976,
      "eval_samples_per_second": 109.38,
      "eval_steps_per_second": 13.673,
      "step": 9000
    },
    {
      "epoch": 1.5058236272878536,
      "grad_norm": 0.41154783964157104,
      "learning_rate": 0.0004999019853491496,
      "loss": 0.4675,
      "step": 9050
    },
    {
      "epoch": 1.51414309484193,
      "grad_norm": 0.5559844374656677,
      "learning_rate": 0.0004999014437120428,
      "loss": 0.4668,
      "step": 9100
    },
    {
      "epoch": 1.5224625623960066,
      "grad_norm": 0.2977549135684967,
      "learning_rate": 0.000499900902074936,
      "loss": 0.4652,
      "step": 9150
    },
    {
      "epoch": 1.5307820299500832,
      "grad_norm": 0.24874624609947205,
      "learning_rate": 0.0004999003604378291,
      "loss": 0.465,
      "step": 9200
    },
    {
      "epoch": 1.5391014975041597,
      "grad_norm": 0.7150341868400574,
      "learning_rate": 0.0004998998188007223,
      "loss": 0.468,
      "step": 9250
    },
    {
      "epoch": 1.5474209650582362,
      "grad_norm": 0.2461651861667633,
      "learning_rate": 0.0004998992771636155,
      "loss": 0.4664,
      "step": 9300
    },
    {
      "epoch": 1.5557404326123128,
      "grad_norm": 0.4250808358192444,
      "learning_rate": 0.0004998987355265087,
      "loss": 0.4641,
      "step": 9350
    },
    {
      "epoch": 1.5640599001663893,
      "grad_norm": 0.3863760530948639,
      "learning_rate": 0.0004998981938894019,
      "loss": 0.4675,
      "step": 9400
    },
    {
      "epoch": 1.5723793677204658,
      "grad_norm": 0.31154608726501465,
      "learning_rate": 0.000499897652252295,
      "loss": 0.4639,
      "step": 9450
    },
    {
      "epoch": 1.5806988352745424,
      "grad_norm": 0.3062475323677063,
      "learning_rate": 0.0004998971106151881,
      "loss": 0.4667,
      "step": 9500
    },
    {
      "epoch": 1.5806988352745424,
      "eval_accuracy": 0.7589258816200106,
      "eval_auc": 0.840721603017736,
      "eval_f1": 0.7787355473290578,
      "eval_loss": 0.46345818042755127,
      "eval_precision": 0.7193166038484428,
      "eval_recall": 0.8488549707670788,
      "eval_runtime": 3071.5791,
      "eval_samples_per_second": 111.31,
      "eval_steps_per_second": 13.914,
      "step": 9500
    },
    {
      "epoch": 1.589018302828619,
      "grad_norm": 0.2855340838432312,
      "learning_rate": 0.0004998965689780813,
      "loss": 0.4682,
      "step": 9550
    },
    {
      "epoch": 1.5973377703826954,
      "grad_norm": 0.309037983417511,
      "learning_rate": 0.0004998960273409745,
      "loss": 0.464,
      "step": 9600
    },
    {
      "epoch": 1.605657237936772,
      "grad_norm": 0.4070012867450714,
      "learning_rate": 0.0004998954857038677,
      "loss": 0.4652,
      "step": 9650
    },
    {
      "epoch": 1.6139767054908485,
      "grad_norm": 0.4464728534221649,
      "learning_rate": 0.0004998949440667609,
      "loss": 0.4655,
      "step": 9700
    },
    {
      "epoch": 1.622296173044925,
      "grad_norm": 0.2548810541629791,
      "learning_rate": 0.000499894402429654,
      "loss": 0.4694,
      "step": 9750
    },
    {
      "epoch": 1.6306156405990015,
      "grad_norm": 0.2812168002128601,
      "learning_rate": 0.0004998938607925473,
      "loss": 0.4618,
      "step": 9800
    },
    {
      "epoch": 1.638935108153078,
      "grad_norm": 0.28069692850112915,
      "learning_rate": 0.0004998933191554404,
      "loss": 0.4669,
      "step": 9850
    },
    {
      "epoch": 1.6472545757071546,
      "grad_norm": 0.2677462100982666,
      "learning_rate": 0.0004998927775183336,
      "loss": 0.469,
      "step": 9900
    },
    {
      "epoch": 1.6555740432612311,
      "grad_norm": 0.3109247386455536,
      "learning_rate": 0.0004998922358812268,
      "loss": 0.4621,
      "step": 9950
    },
    {
      "epoch": 1.6638935108153077,
      "grad_norm": 0.3598281741142273,
      "learning_rate": 0.0004998916942441199,
      "loss": 0.4614,
      "step": 10000
    },
    {
      "epoch": 1.6638935108153077,
      "eval_accuracy": 0.7620525359828487,
      "eval_auc": 0.8427547777707038,
      "eval_f1": 0.7841862883458367,
      "eval_loss": 0.460246205329895,
      "eval_precision": 0.7171692666003542,
      "eval_recall": 0.8650194008275346,
      "eval_runtime": 3091.9645,
      "eval_samples_per_second": 110.577,
      "eval_steps_per_second": 13.822,
      "step": 10000
    },
    {
      "epoch": 1.6722129783693842,
      "grad_norm": 0.5699976682662964,
      "learning_rate": 0.0004998911526070131,
      "loss": 0.4668,
      "step": 10050
    },
    {
      "epoch": 1.6805324459234607,
      "grad_norm": 0.3142874538898468,
      "learning_rate": 0.0004998906109699063,
      "loss": 0.4634,
      "step": 10100
    },
    {
      "epoch": 1.6888519134775375,
      "grad_norm": 0.3907163441181183,
      "learning_rate": 0.0004998900693327995,
      "loss": 0.4686,
      "step": 10150
    },
    {
      "epoch": 1.697171381031614,
      "grad_norm": 0.39660388231277466,
      "learning_rate": 0.0004998895276956927,
      "loss": 0.465,
      "step": 10200
    },
    {
      "epoch": 1.7054908485856906,
      "grad_norm": 0.6148430705070496,
      "learning_rate": 0.0004998889860585859,
      "loss": 0.4654,
      "step": 10250
    },
    {
      "epoch": 1.713810316139767,
      "grad_norm": 0.2848416864871979,
      "learning_rate": 0.000499888444421479,
      "loss": 0.4617,
      "step": 10300
    },
    {
      "epoch": 1.7221297836938436,
      "grad_norm": 0.4551065266132355,
      "learning_rate": 0.0004998879027843723,
      "loss": 0.4622,
      "step": 10350
    },
    {
      "epoch": 1.7304492512479202,
      "grad_norm": 0.24504904448986053,
      "learning_rate": 0.0004998873611472654,
      "loss": 0.4558,
      "step": 10400
    },
    {
      "epoch": 1.7387687188019967,
      "grad_norm": 0.2634543776512146,
      "learning_rate": 0.0004998868195101586,
      "loss": 0.4621,
      "step": 10450
    },
    {
      "epoch": 1.7470881863560732,
      "grad_norm": 0.3404432237148285,
      "learning_rate": 0.0004998862778730518,
      "loss": 0.465,
      "step": 10500
    },
    {
      "epoch": 1.7470881863560732,
      "eval_accuracy": 0.7618653461987313,
      "eval_auc": 0.8441348324318876,
      "eval_f1": 0.7790148521301081,
      "eval_loss": 0.45925500988960266,
      "eval_precision": 0.7263860135754164,
      "eval_recall": 0.8398656280542404,
      "eval_runtime": 3075.9859,
      "eval_samples_per_second": 111.151,
      "eval_steps_per_second": 13.894,
      "step": 10500
    },
    {
      "epoch": 1.7554076539101497,
      "grad_norm": 0.35343432426452637,
      "learning_rate": 0.0004998857362359449,
      "loss": 0.4592,
      "step": 10550
    },
    {
      "epoch": 1.7637271214642263,
      "grad_norm": 0.3516612648963928,
      "learning_rate": 0.0004998851945988382,
      "loss": 0.464,
      "step": 10600
    },
    {
      "epoch": 1.7720465890183028,
      "grad_norm": 0.295074462890625,
      "learning_rate": 0.0004998846529617313,
      "loss": 0.4642,
      "step": 10650
    },
    {
      "epoch": 1.7803660565723793,
      "grad_norm": 0.3098844885826111,
      "learning_rate": 0.0004998841113246245,
      "loss": 0.463,
      "step": 10700
    },
    {
      "epoch": 1.7886855241264559,
      "grad_norm": 0.3475278317928314,
      "learning_rate": 0.0004998835696875177,
      "loss": 0.4593,
      "step": 10750
    },
    {
      "epoch": 1.7970049916805324,
      "grad_norm": 0.2945404350757599,
      "learning_rate": 0.0004998830280504109,
      "loss": 0.4663,
      "step": 10800
    },
    {
      "epoch": 1.805324459234609,
      "grad_norm": 0.3487035632133484,
      "learning_rate": 0.000499882486413304,
      "loss": 0.4637,
      "step": 10850
    },
    {
      "epoch": 1.8136439267886857,
      "grad_norm": 0.3474220931529999,
      "learning_rate": 0.0004998819447761971,
      "loss": 0.4639,
      "step": 10900
    },
    {
      "epoch": 1.8219633943427622,
      "grad_norm": 0.5383617877960205,
      "learning_rate": 0.0004998814031390903,
      "loss": 0.4615,
      "step": 10950
    },
    {
      "epoch": 1.8302828618968388,
      "grad_norm": 0.30488741397857666,
      "learning_rate": 0.0004998808615019835,
      "loss": 0.4602,
      "step": 11000
    },
    {
      "epoch": 1.8302828618968388,
      "eval_accuracy": 0.7627603473540432,
      "eval_auc": 0.8444230596164721,
      "eval_f1": 0.7833894140896224,
      "eval_loss": 0.4582211673259735,
      "eval_precision": 0.720434596814201,
      "eval_recall": 0.8584002949628078,
      "eval_runtime": 3088.1236,
      "eval_samples_per_second": 110.714,
      "eval_steps_per_second": 13.839,
      "step": 11000
    },
    {
      "epoch": 1.8386023294509153,
      "grad_norm": 0.3203362822532654,
      "learning_rate": 0.0004998803198648767,
      "loss": 0.461,
      "step": 11050
    },
    {
      "epoch": 1.8469217970049918,
      "grad_norm": 0.272316038608551,
      "learning_rate": 0.0004998797782277699,
      "loss": 0.4634,
      "step": 11100
    },
    {
      "epoch": 1.8552412645590683,
      "grad_norm": 0.3532938063144684,
      "learning_rate": 0.0004998792365906631,
      "loss": 0.4616,
      "step": 11150
    },
    {
      "epoch": 1.8635607321131449,
      "grad_norm": 0.32843783497810364,
      "learning_rate": 0.0004998786949535562,
      "loss": 0.4698,
      "step": 11200
    },
    {
      "epoch": 1.8718801996672214,
      "grad_norm": 0.43027517199516296,
      "learning_rate": 0.0004998781533164494,
      "loss": 0.4561,
      "step": 11250
    },
    {
      "epoch": 1.880199667221298,
      "grad_norm": 0.7312646508216858,
      "learning_rate": 0.0004998776116793426,
      "loss": 0.4621,
      "step": 11300
    },
    {
      "epoch": 1.8885191347753745,
      "grad_norm": 0.33167243003845215,
      "learning_rate": 0.0004998770700422358,
      "loss": 0.4587,
      "step": 11350
    },
    {
      "epoch": 1.896838602329451,
      "grad_norm": 0.25895464420318604,
      "learning_rate": 0.000499876528405129,
      "loss": 0.4629,
      "step": 11400
    },
    {
      "epoch": 1.9051580698835275,
      "grad_norm": 0.34123238921165466,
      "learning_rate": 0.0004998759867680221,
      "loss": 0.4603,
      "step": 11450
    },
    {
      "epoch": 1.913477537437604,
      "grad_norm": 0.3420165479183197,
      "learning_rate": 0.0004998754451309153,
      "loss": 0.4611,
      "step": 11500
    },
    {
      "epoch": 1.913477537437604,
      "eval_accuracy": 0.7647872617351908,
      "eval_auc": 0.8470155046356656,
      "eval_f1": 0.7896492104554169,
      "eval_loss": 0.4563252329826355,
      "eval_precision": 0.7138904653802497,
      "eval_recall": 0.8833960519462278,
      "eval_runtime": 3116.3735,
      "eval_samples_per_second": 109.711,
      "eval_steps_per_second": 13.714,
      "step": 11500
    },
    {
      "epoch": 1.9217970049916806,
      "grad_norm": 0.31535664200782776,
      "learning_rate": 0.0004998749034938085,
      "loss": 0.454,
      "step": 11550
    },
    {
      "epoch": 1.9301164725457571,
      "grad_norm": 0.3028728663921356,
      "learning_rate": 0.0004998743618567017,
      "loss": 0.4655,
      "step": 11600
    },
    {
      "epoch": 1.9384359400998337,
      "grad_norm": 0.35277822613716125,
      "learning_rate": 0.0004998738202195949,
      "loss": 0.4579,
      "step": 11650
    },
    {
      "epoch": 1.9467554076539102,
      "grad_norm": 0.2921321988105774,
      "learning_rate": 0.0004998732785824881,
      "loss": 0.4597,
      "step": 11700
    },
    {
      "epoch": 1.9550748752079867,
      "grad_norm": 0.49088171124458313,
      "learning_rate": 0.0004998727369453812,
      "loss": 0.454,
      "step": 11750
    },
    {
      "epoch": 1.9633943427620633,
      "grad_norm": 0.37039515376091003,
      "learning_rate": 0.0004998721953082744,
      "loss": 0.463,
      "step": 11800
    },
    {
      "epoch": 1.9717138103161398,
      "grad_norm": 0.3478553295135498,
      "learning_rate": 0.0004998716536711676,
      "loss": 0.4596,
      "step": 11850
    },
    {
      "epoch": 1.9800332778702163,
      "grad_norm": 0.31678304076194763,
      "learning_rate": 0.0004998711120340608,
      "loss": 0.4627,
      "step": 11900
    },
    {
      "epoch": 1.9883527454242929,
      "grad_norm": 0.2272387593984604,
      "learning_rate": 0.000499870570396954,
      "loss": 0.4627,
      "step": 11950
    },
    {
      "epoch": 1.9966722129783694,
      "grad_norm": 0.26735177636146545,
      "learning_rate": 0.0004998700287598471,
      "loss": 0.4614,
      "step": 12000
    },
    {
      "epoch": 1.9966722129783694,
      "eval_accuracy": 0.7641379471715332,
      "eval_auc": 0.8468346428705298,
      "eval_f1": 0.7830218885794621,
      "eval_loss": 0.45656195282936096,
      "eval_precision": 0.72468698016794,
      "eval_recall": 0.8515705013782489,
      "eval_runtime": 3066.5927,
      "eval_samples_per_second": 111.491,
      "eval_steps_per_second": 13.937,
      "step": 12000
    },
    {
      "epoch": 2.004991680532446,
      "grad_norm": 0.4045884907245636,
      "learning_rate": 0.0004998694871227403,
      "loss": 0.457,
      "step": 12050
    },
    {
      "epoch": 2.0133111480865225,
      "grad_norm": 0.3358103036880493,
      "learning_rate": 0.0004998689454856335,
      "loss": 0.4457,
      "step": 12100
    },
    {
      "epoch": 2.021630615640599,
      "grad_norm": 0.4295903742313385,
      "learning_rate": 0.0004998684038485267,
      "loss": 0.4526,
      "step": 12150
    },
    {
      "epoch": 2.0299500831946755,
      "grad_norm": 0.30707797408103943,
      "learning_rate": 0.0004998678622114199,
      "loss": 0.4523,
      "step": 12200
    },
    {
      "epoch": 2.038269550748752,
      "grad_norm": 0.30385756492614746,
      "learning_rate": 0.000499867320574313,
      "loss": 0.4529,
      "step": 12250
    },
    {
      "epoch": 2.0465890183028286,
      "grad_norm": 0.25600412487983704,
      "learning_rate": 0.0004998667897699483,
      "loss": 0.4551,
      "step": 12300
    },
    {
      "epoch": 2.054908485856905,
      "grad_norm": 0.4305065870285034,
      "learning_rate": 0.0004998662481328415,
      "loss": 0.4545,
      "step": 12350
    },
    {
      "epoch": 2.0632279534109816,
      "grad_norm": 0.49212226271629333,
      "learning_rate": 0.0004998657064957347,
      "loss": 0.4561,
      "step": 12400
    },
    {
      "epoch": 2.071547420965058,
      "grad_norm": 0.2785611152648926,
      "learning_rate": 0.0004998651648586279,
      "loss": 0.457,
      "step": 12450
    },
    {
      "epoch": 2.0798668885191347,
      "grad_norm": 0.37156569957733154,
      "learning_rate": 0.0004998646232215211,
      "loss": 0.4543,
      "step": 12500
    },
    {
      "epoch": 2.0798668885191347,
      "eval_accuracy": 0.7628480925653482,
      "eval_auc": 0.847242187715394,
      "eval_f1": 0.791982226053404,
      "eval_loss": 0.45966818928718567,
      "eval_precision": 0.7050727451293881,
      "eval_recall": 0.9033294512170142,
      "eval_runtime": 3110.1977,
      "eval_samples_per_second": 109.928,
      "eval_steps_per_second": 13.741,
      "step": 12500
    },
    {
      "epoch": 2.0881863560732112,
      "grad_norm": 0.3126935660839081,
      "learning_rate": 0.0004998640815844142,
      "loss": 0.4533,
      "step": 12550
    },
    {
      "epoch": 2.0965058236272878,
      "grad_norm": 0.2464996725320816,
      "learning_rate": 0.0004998635399473074,
      "loss": 0.4572,
      "step": 12600
    },
    {
      "epoch": 2.1048252911813643,
      "grad_norm": 0.5269837379455566,
      "learning_rate": 0.0004998629983102006,
      "loss": 0.4573,
      "step": 12650
    },
    {
      "epoch": 2.113144758735441,
      "grad_norm": 0.3495631217956543,
      "learning_rate": 0.0004998624566730938,
      "loss": 0.4509,
      "step": 12700
    },
    {
      "epoch": 2.1214642262895174,
      "grad_norm": 0.24060170352458954,
      "learning_rate": 0.000499861915035987,
      "loss": 0.4542,
      "step": 12750
    },
    {
      "epoch": 2.129783693843594,
      "grad_norm": 0.43895021080970764,
      "learning_rate": 0.0004998613733988801,
      "loss": 0.4528,
      "step": 12800
    },
    {
      "epoch": 2.1381031613976704,
      "grad_norm": 0.30949124693870544,
      "learning_rate": 0.0004998608317617733,
      "loss": 0.4481,
      "step": 12850
    },
    {
      "epoch": 2.146422628951747,
      "grad_norm": 0.3358418345451355,
      "learning_rate": 0.0004998602901246665,
      "loss": 0.4504,
      "step": 12900
    },
    {
      "epoch": 2.1547420965058235,
      "grad_norm": 0.40767422318458557,
      "learning_rate": 0.0004998597484875597,
      "loss": 0.4583,
      "step": 12950
    },
    {
      "epoch": 2.1630615640599,
      "grad_norm": 0.3742935061454773,
      "learning_rate": 0.0004998592068504528,
      "loss": 0.4574,
      "step": 13000
    },
    {
      "epoch": 2.1630615640599,
      "eval_accuracy": 0.7631610504856697,
      "eval_auc": 0.8489359071807063,
      "eval_f1": 0.794699065212045,
      "eval_loss": 0.45762190222740173,
      "eval_precision": 0.7010583667334669,
      "eval_recall": 0.9172114309792883,
      "eval_runtime": 3073.781,
      "eval_samples_per_second": 111.231,
      "eval_steps_per_second": 13.904,
      "step": 13000
    },
    {
      "epoch": 2.1713810316139766,
      "grad_norm": 0.3705851137638092,
      "learning_rate": 0.000499858665213346,
      "loss": 0.4513,
      "step": 13050
    },
    {
      "epoch": 2.179700499168053,
      "grad_norm": 0.222339928150177,
      "learning_rate": 0.0004998581235762391,
      "loss": 0.4536,
      "step": 13100
    },
    {
      "epoch": 2.1880199667221296,
      "grad_norm": 0.3179910182952881,
      "learning_rate": 0.0004998575819391323,
      "loss": 0.4527,
      "step": 13150
    },
    {
      "epoch": 2.196339434276206,
      "grad_norm": 0.43729254603385925,
      "learning_rate": 0.0004998570403020255,
      "loss": 0.4521,
      "step": 13200
    },
    {
      "epoch": 2.2046589018302827,
      "grad_norm": 0.31573063135147095,
      "learning_rate": 0.0004998564986649187,
      "loss": 0.4573,
      "step": 13250
    },
    {
      "epoch": 2.212978369384359,
      "grad_norm": 0.34133201837539673,
      "learning_rate": 0.0004998559570278119,
      "loss": 0.4547,
      "step": 13300
    },
    {
      "epoch": 2.2212978369384357,
      "grad_norm": 0.32229501008987427,
      "learning_rate": 0.000499855415390705,
      "loss": 0.4552,
      "step": 13350
    },
    {
      "epoch": 2.2296173044925123,
      "grad_norm": 0.25182995200157166,
      "learning_rate": 0.0004998548737535982,
      "loss": 0.4519,
      "step": 13400
    },
    {
      "epoch": 2.237936772046589,
      "grad_norm": 0.38632023334503174,
      "learning_rate": 0.0004998543321164914,
      "loss": 0.4549,
      "step": 13450
    },
    {
      "epoch": 2.2462562396006653,
      "grad_norm": 0.4629928469657898,
      "learning_rate": 0.0004998537904793846,
      "loss": 0.4532,
      "step": 13500
    },
    {
      "epoch": 2.2462562396006653,
      "eval_accuracy": 0.7667761531914395,
      "eval_auc": 0.8491075200735866,
      "eval_f1": 0.7892359376528085,
      "eval_loss": 0.4522622525691986,
      "eval_precision": 0.7196284656615124,
      "eval_recall": 0.8737512363272448,
      "eval_runtime": 3076.782,
      "eval_samples_per_second": 111.122,
      "eval_steps_per_second": 13.89,
      "step": 13500
    },
    {
      "epoch": 2.254575707154742,
      "grad_norm": 0.3204486668109894,
      "learning_rate": 0.0004998532488422778,
      "loss": 0.4549,
      "step": 13550
    },
    {
      "epoch": 2.2628951747088184,
      "grad_norm": 0.2668420076370239,
      "learning_rate": 0.0004998527072051709,
      "loss": 0.4498,
      "step": 13600
    },
    {
      "epoch": 2.271214642262895,
      "grad_norm": 0.34950193762779236,
      "learning_rate": 0.0004998521655680641,
      "loss": 0.4492,
      "step": 13650
    },
    {
      "epoch": 2.2795341098169715,
      "grad_norm": 0.39468878507614136,
      "learning_rate": 0.0004998516239309573,
      "loss": 0.4572,
      "step": 13700
    },
    {
      "epoch": 2.2878535773710484,
      "grad_norm": 0.40462568402290344,
      "learning_rate": 0.0004998510822938505,
      "loss": 0.4521,
      "step": 13750
    },
    {
      "epoch": 2.296173044925125,
      "grad_norm": 0.3331548571586609,
      "learning_rate": 0.0004998505406567437,
      "loss": 0.4473,
      "step": 13800
    },
    {
      "epoch": 2.3044925124792015,
      "grad_norm": 0.2660551965236664,
      "learning_rate": 0.0004998499990196369,
      "loss": 0.4534,
      "step": 13850
    },
    {
      "epoch": 2.312811980033278,
      "grad_norm": 0.311027854681015,
      "learning_rate": 0.00049984945738253,
      "loss": 0.45,
      "step": 13900
    },
    {
      "epoch": 2.3211314475873546,
      "grad_norm": 0.5009548664093018,
      "learning_rate": 0.0004998489157454232,
      "loss": 0.4558,
      "step": 13950
    },
    {
      "epoch": 2.329450915141431,
      "grad_norm": 0.28578829765319824,
      "learning_rate": 0.0004998483741083164,
      "loss": 0.4472,
      "step": 14000
    },
    {
      "epoch": 2.329450915141431,
      "eval_accuracy": 0.7664251723462192,
      "eval_auc": 0.8494449173929353,
      "eval_f1": 0.7848782550838166,
      "eval_loss": 0.452217161655426,
      "eval_precision": 0.7271184579602511,
      "eval_recall": 0.8526063826674236,
      "eval_runtime": 3063.8475,
      "eval_samples_per_second": 111.591,
      "eval_steps_per_second": 13.949,
      "step": 14000
    },
    {
      "epoch": 2.3377703826955076,
      "grad_norm": 0.351082444190979,
      "learning_rate": 0.0004998478324712096,
      "loss": 0.4548,
      "step": 14050
    },
    {
      "epoch": 2.346089850249584,
      "grad_norm": 0.35973915457725525,
      "learning_rate": 0.0004998472908341028,
      "loss": 0.4514,
      "step": 14100
    },
    {
      "epoch": 2.3544093178036607,
      "grad_norm": 0.4456665515899658,
      "learning_rate": 0.0004998467491969959,
      "loss": 0.4553,
      "step": 14150
    },
    {
      "epoch": 2.3627287853577372,
      "grad_norm": 0.34595686197280884,
      "learning_rate": 0.0004998462075598891,
      "loss": 0.4559,
      "step": 14200
    },
    {
      "epoch": 2.3710482529118138,
      "grad_norm": 0.3657195270061493,
      "learning_rate": 0.0004998456659227823,
      "loss": 0.4499,
      "step": 14250
    },
    {
      "epoch": 2.3793677204658903,
      "grad_norm": 0.398843377828598,
      "learning_rate": 0.0004998451242856755,
      "loss": 0.4523,
      "step": 14300
    },
    {
      "epoch": 2.387687188019967,
      "grad_norm": 0.44224244356155396,
      "learning_rate": 0.0004998445826485687,
      "loss": 0.4479,
      "step": 14350
    },
    {
      "epoch": 2.3960066555740434,
      "grad_norm": 0.4334554374217987,
      "learning_rate": 0.0004998440410114619,
      "loss": 0.452,
      "step": 14400
    },
    {
      "epoch": 2.40432612312812,
      "grad_norm": 0.36450016498565674,
      "learning_rate": 0.0004998434993743549,
      "loss": 0.4517,
      "step": 14450
    },
    {
      "epoch": 2.4126455906821964,
      "grad_norm": 0.37175169587135315,
      "learning_rate": 0.0004998429577372481,
      "loss": 0.4505,
      "step": 14500
    },
    {
      "epoch": 2.4126455906821964,
      "eval_accuracy": 0.7648048107774518,
      "eval_auc": 0.8497144763208656,
      "eval_f1": 0.7768245941722891,
      "eval_loss": 0.45364442467689514,
      "eval_precision": 0.7387407360486477,
      "eval_recall": 0.8190485108474913,
      "eval_runtime": 3069.7075,
      "eval_samples_per_second": 111.378,
      "eval_steps_per_second": 13.922,
      "step": 14500
    },
    {
      "epoch": 2.420965058236273,
      "grad_norm": 0.4115402102470398,
      "learning_rate": 0.0004998424161001413,
      "loss": 0.4522,
      "step": 14550
    },
    {
      "epoch": 2.4292845257903495,
      "grad_norm": 0.33581915497779846,
      "learning_rate": 0.0004998418744630345,
      "loss": 0.4546,
      "step": 14600
    },
    {
      "epoch": 2.437603993344426,
      "grad_norm": 0.26651909947395325,
      "learning_rate": 0.0004998413328259277,
      "loss": 0.4527,
      "step": 14650
    },
    {
      "epoch": 2.4459234608985025,
      "grad_norm": 0.3490288555622101,
      "learning_rate": 0.0004998407911888208,
      "loss": 0.4479,
      "step": 14700
    },
    {
      "epoch": 2.454242928452579,
      "grad_norm": 0.3200080990791321,
      "learning_rate": 0.000499840249551714,
      "loss": 0.4471,
      "step": 14750
    },
    {
      "epoch": 2.4625623960066556,
      "grad_norm": 0.31173649430274963,
      "learning_rate": 0.0004998397079146072,
      "loss": 0.4488,
      "step": 14800
    },
    {
      "epoch": 2.470881863560732,
      "grad_norm": 0.431302934885025,
      "learning_rate": 0.0004998391662775004,
      "loss": 0.4509,
      "step": 14850
    },
    {
      "epoch": 2.4792013311148087,
      "grad_norm": 0.37330421805381775,
      "learning_rate": 0.0004998386246403936,
      "loss": 0.4537,
      "step": 14900
    },
    {
      "epoch": 2.487520798668885,
      "grad_norm": 0.2857615649700165,
      "learning_rate": 0.0004998380830032867,
      "loss": 0.4469,
      "step": 14950
    },
    {
      "epoch": 2.4958402662229617,
      "grad_norm": 0.3119502365589142,
      "learning_rate": 0.0004998375413661799,
      "loss": 0.4506,
      "step": 15000
    },
    {
      "epoch": 2.4958402662229617,
      "eval_accuracy": 0.7662789303273774,
      "eval_auc": 0.8515752861892085,
      "eval_f1": 0.7784029616893832,
      "eval_loss": 0.45170241594314575,
      "eval_precision": 0.7397014799510899,
      "eval_recall": 0.821377780638969,
      "eval_runtime": 3072.2397,
      "eval_samples_per_second": 111.287,
      "eval_steps_per_second": 13.911,
      "step": 15000
    },
    {
      "epoch": 2.5041597337770383,
      "grad_norm": 0.31348904967308044,
      "learning_rate": 0.0004998369997290731,
      "loss": 0.4502,
      "step": 15050
    },
    {
      "epoch": 2.512479201331115,
      "grad_norm": 0.280769020318985,
      "learning_rate": 0.0004998364580919663,
      "loss": 0.4507,
      "step": 15100
    },
    {
      "epoch": 2.5207986688851913,
      "grad_norm": 0.3579796850681305,
      "learning_rate": 0.0004998359164548595,
      "loss": 0.45,
      "step": 15150
    },
    {
      "epoch": 2.529118136439268,
      "grad_norm": 0.4033232629299164,
      "learning_rate": 0.0004998353748177527,
      "loss": 0.4506,
      "step": 15200
    },
    {
      "epoch": 2.5374376039933444,
      "grad_norm": 0.3025188148021698,
      "learning_rate": 0.0004998348331806458,
      "loss": 0.4515,
      "step": 15250
    },
    {
      "epoch": 2.545757071547421,
      "grad_norm": 0.28588080406188965,
      "learning_rate": 0.0004998342915435391,
      "loss": 0.4502,
      "step": 15300
    },
    {
      "epoch": 2.5540765391014975,
      "grad_norm": 0.36141297221183777,
      "learning_rate": 0.0004998337499064322,
      "loss": 0.446,
      "step": 15350
    },
    {
      "epoch": 2.562396006655574,
      "grad_norm": 0.5286752581596375,
      "learning_rate": 0.0004998332082693254,
      "loss": 0.4511,
      "step": 15400
    },
    {
      "epoch": 2.5707154742096505,
      "grad_norm": 0.3852572441101074,
      "learning_rate": 0.0004998326666322186,
      "loss": 0.455,
      "step": 15450
    },
    {
      "epoch": 2.579034941763727,
      "grad_norm": 0.3271165192127228,
      "learning_rate": 0.0004998321249951117,
      "loss": 0.4562,
      "step": 15500
    },
    {
      "epoch": 2.579034941763727,
      "eval_accuracy": 0.7698033629814652,
      "eval_auc": 0.8525774850851346,
      "eval_f1": 0.7916979853693137,
      "eval_loss": 0.447857141494751,
      "eval_precision": 0.7226562688737818,
      "eval_recall": 0.875325541789324,
      "eval_runtime": 3063.8515,
      "eval_samples_per_second": 111.591,
      "eval_steps_per_second": 13.949,
      "step": 15500
    },
    {
      "epoch": 2.5873544093178036,
      "grad_norm": 0.29755738377571106,
      "learning_rate": 0.000499831583358005,
      "loss": 0.4446,
      "step": 15550
    },
    {
      "epoch": 2.59567387687188,
      "grad_norm": 0.3373735547065735,
      "learning_rate": 0.0004998310417208981,
      "loss": 0.4515,
      "step": 15600
    },
    {
      "epoch": 2.6039933444259566,
      "grad_norm": 0.33765482902526855,
      "learning_rate": 0.0004998305000837913,
      "loss": 0.4488,
      "step": 15650
    },
    {
      "epoch": 2.612312811980033,
      "grad_norm": 0.4411982595920563,
      "learning_rate": 0.0004998299584466845,
      "loss": 0.4498,
      "step": 15700
    },
    {
      "epoch": 2.6206322795341097,
      "grad_norm": 0.3372790217399597,
      "learning_rate": 0.0004998294168095777,
      "loss": 0.4496,
      "step": 15750
    },
    {
      "epoch": 2.6289517470881862,
      "grad_norm": 0.2755867540836334,
      "learning_rate": 0.0004998288751724709,
      "loss": 0.4504,
      "step": 15800
    },
    {
      "epoch": 2.6372712146422628,
      "grad_norm": 0.33905473351478577,
      "learning_rate": 0.0004998283335353639,
      "loss": 0.4549,
      "step": 15850
    },
    {
      "epoch": 2.6455906821963393,
      "grad_norm": 0.38283759355545044,
      "learning_rate": 0.0004998277918982571,
      "loss": 0.445,
      "step": 15900
    },
    {
      "epoch": 2.653910149750416,
      "grad_norm": 0.3542182743549347,
      "learning_rate": 0.0004998272502611503,
      "loss": 0.448,
      "step": 15950
    },
    {
      "epoch": 2.6622296173044924,
      "grad_norm": 0.42483657598495483,
      "learning_rate": 0.0004998267086240435,
      "loss": 0.4434,
      "step": 16000
    },
    {
      "epoch": 2.6622296173044924,
      "eval_accuracy": 0.769505029263028,
      "eval_auc": 0.8523741424932588,
      "eval_f1": 0.7897149627226103,
      "eval_loss": 0.4486808776855469,
      "eval_precision": 0.7257674518978464,
      "eval_recall": 0.8660201674967373,
      "eval_runtime": 3076.2646,
      "eval_samples_per_second": 111.141,
      "eval_steps_per_second": 13.893,
      "step": 16000
    },
    {
      "epoch": 2.670549084858569,
      "grad_norm": 0.2930859327316284,
      "learning_rate": 0.0004998261669869367,
      "loss": 0.4467,
      "step": 16050
    },
    {
      "epoch": 2.6788685524126454,
      "grad_norm": 0.4802510142326355,
      "learning_rate": 0.0004998256253498299,
      "loss": 0.4472,
      "step": 16100
    },
    {
      "epoch": 2.687188019966722,
      "grad_norm": 0.30784839391708374,
      "learning_rate": 0.000499825083712723,
      "loss": 0.4489,
      "step": 16150
    },
    {
      "epoch": 2.695507487520799,
      "grad_norm": 0.27913177013397217,
      "learning_rate": 0.0004998245420756162,
      "loss": 0.451,
      "step": 16200
    },
    {
      "epoch": 2.7038269550748755,
      "grad_norm": 0.3060939311981201,
      "learning_rate": 0.0004998240004385094,
      "loss": 0.446,
      "step": 16250
    },
    {
      "epoch": 2.712146422628952,
      "grad_norm": 0.3916643261909485,
      "learning_rate": 0.0004998234696341447,
      "loss": 0.449,
      "step": 16300
    },
    {
      "epoch": 2.7204658901830285,
      "grad_norm": 0.3149168789386749,
      "learning_rate": 0.0004998229279970379,
      "loss": 0.4484,
      "step": 16350
    },
    {
      "epoch": 2.728785357737105,
      "grad_norm": 0.3204551339149475,
      "learning_rate": 0.000499822386359931,
      "loss": 0.4467,
      "step": 16400
    },
    {
      "epoch": 2.7371048252911816,
      "grad_norm": 0.4137204885482788,
      "learning_rate": 0.0004998218447228243,
      "loss": 0.4503,
      "step": 16450
    },
    {
      "epoch": 2.745424292845258,
      "grad_norm": 0.34219491481781006,
      "learning_rate": 0.0004998213030857175,
      "loss": 0.4482,
      "step": 16500
    },
    {
      "epoch": 2.745424292845258,
      "eval_accuracy": 0.7704673017470072,
      "eval_auc": 0.8527454649367189,
      "eval_f1": 0.7930540722281555,
      "eval_loss": 0.4480786919593811,
      "eval_precision": 0.7217273189790061,
      "eval_recall": 0.8800250484289134,
      "eval_runtime": 3081.5516,
      "eval_samples_per_second": 110.95,
      "eval_steps_per_second": 13.869,
      "step": 16500
    },
    {
      "epoch": 2.7537437603993347,
      "grad_norm": 0.38494977355003357,
      "learning_rate": 0.0004998207614486106,
      "loss": 0.4491,
      "step": 16550
    },
    {
      "epoch": 2.762063227953411,
      "grad_norm": 0.3267413079738617,
      "learning_rate": 0.0004998202198115037,
      "loss": 0.4502,
      "step": 16600
    },
    {
      "epoch": 2.7703826955074877,
      "grad_norm": 0.39887621998786926,
      "learning_rate": 0.000499819678174397,
      "loss": 0.4529,
      "step": 16650
    },
    {
      "epoch": 2.7787021630615643,
      "grad_norm": 0.4172428548336029,
      "learning_rate": 0.0004998191365372901,
      "loss": 0.4517,
      "step": 16700
    },
    {
      "epoch": 2.787021630615641,
      "grad_norm": 0.3547029197216034,
      "learning_rate": 0.0004998185949001833,
      "loss": 0.4471,
      "step": 16750
    },
    {
      "epoch": 2.7953410981697173,
      "grad_norm": 0.31628596782684326,
      "learning_rate": 0.0004998180532630765,
      "loss": 0.4472,
      "step": 16800
    },
    {
      "epoch": 2.803660565723794,
      "grad_norm": 0.29753875732421875,
      "learning_rate": 0.0004998175116259696,
      "loss": 0.4506,
      "step": 16850
    },
    {
      "epoch": 2.8119800332778704,
      "grad_norm": 0.48493313789367676,
      "learning_rate": 0.0004998169699888629,
      "loss": 0.4502,
      "step": 16900
    },
    {
      "epoch": 2.820299500831947,
      "grad_norm": 0.36175137758255005,
      "learning_rate": 0.000499816428351756,
      "loss": 0.4504,
      "step": 16950
    },
    {
      "epoch": 2.8286189683860234,
      "grad_norm": 0.3172094523906708,
      "learning_rate": 0.0004998158867146492,
      "loss": 0.4455,
      "step": 17000
    },
    {
      "epoch": 2.8286189683860234,
      "eval_accuracy": 0.770005176967467,
      "eval_auc": 0.85512334595064,
      "eval_f1": 0.782045916288558,
      "eval_loss": 0.4468492567539215,
      "eval_precision": 0.7428258511568151,
      "eval_recall": 0.8256383545289081,
      "eval_runtime": 3103.5745,
      "eval_samples_per_second": 110.163,
      "eval_steps_per_second": 13.771,
      "step": 17000
    },
    {
      "epoch": 2.8369384359401,
      "grad_norm": 0.49940770864486694,
      "learning_rate": 0.0004998153450775424,
      "loss": 0.4467,
      "step": 17050
    },
    {
      "epoch": 2.8452579034941765,
      "grad_norm": 0.3779182434082031,
      "learning_rate": 0.0004998148034404356,
      "loss": 0.4427,
      "step": 17100
    },
    {
      "epoch": 2.853577371048253,
      "grad_norm": 0.3529649078845978,
      "learning_rate": 0.0004998142726360708,
      "loss": 0.4469,
      "step": 17150
    },
    {
      "epoch": 2.8618968386023296,
      "grad_norm": 0.421783447265625,
      "learning_rate": 0.0004998137309989642,
      "loss": 0.4449,
      "step": 17200
    },
    {
      "epoch": 2.870216306156406,
      "grad_norm": 0.38324880599975586,
      "learning_rate": 0.0004998131893618572,
      "loss": 0.4446,
      "step": 17250
    },
    {
      "epoch": 2.8785357737104826,
      "grad_norm": 0.43824031949043274,
      "learning_rate": 0.0004998126477247504,
      "loss": 0.4465,
      "step": 17300
    },
    {
      "epoch": 2.886855241264559,
      "grad_norm": 0.3818030059337616,
      "learning_rate": 0.0004998121060876436,
      "loss": 0.4459,
      "step": 17350
    },
    {
      "epoch": 2.8951747088186357,
      "grad_norm": 0.30165761709213257,
      "learning_rate": 0.0004998115644505367,
      "loss": 0.4505,
      "step": 17400
    },
    {
      "epoch": 2.9034941763727122,
      "grad_norm": 0.28733378648757935,
      "learning_rate": 0.00049981102281343,
      "loss": 0.4526,
      "step": 17450
    },
    {
      "epoch": 2.9118136439267888,
      "grad_norm": 0.3203652799129486,
      "learning_rate": 0.0004998104811763231,
      "loss": 0.4428,
      "step": 17500
    },
    {
      "epoch": 2.9118136439267888,
      "eval_accuracy": 0.7715377933249293,
      "eval_auc": 0.8548695984738128,
      "eval_f1": 0.7888806302958228,
      "eval_loss": 0.44894230365753174,
      "eval_precision": 0.7329245264067177,
      "eval_recall": 0.8540870491429107,
      "eval_runtime": 3097.7551,
      "eval_samples_per_second": 110.37,
      "eval_steps_per_second": 13.796,
      "step": 17500
    },
    {
      "epoch": 2.9201331114808653,
      "grad_norm": 0.2848743498325348,
      "learning_rate": 0.0004998099395392163,
      "loss": 0.4506,
      "step": 17550
    },
    {
      "epoch": 2.928452579034942,
      "grad_norm": 0.3639114499092102,
      "learning_rate": 0.0004998093979021095,
      "loss": 0.45,
      "step": 17600
    },
    {
      "epoch": 2.9367720465890184,
      "grad_norm": 0.35603734850883484,
      "learning_rate": 0.0004998088562650027,
      "loss": 0.4445,
      "step": 17650
    },
    {
      "epoch": 2.945091514143095,
      "grad_norm": 0.3066865801811218,
      "learning_rate": 0.0004998083146278958,
      "loss": 0.4486,
      "step": 17700
    },
    {
      "epoch": 2.9534109816971714,
      "grad_norm": 0.30339953303337097,
      "learning_rate": 0.000499807772990789,
      "loss": 0.4523,
      "step": 17750
    },
    {
      "epoch": 2.961730449251248,
      "grad_norm": 0.2876887917518616,
      "learning_rate": 0.0004998072313536822,
      "loss": 0.4506,
      "step": 17800
    },
    {
      "epoch": 2.9700499168053245,
      "grad_norm": 0.32773932814598083,
      "learning_rate": 0.0004998066897165754,
      "loss": 0.4459,
      "step": 17850
    },
    {
      "epoch": 2.978369384359401,
      "grad_norm": 0.2950334846973419,
      "learning_rate": 0.0004998061480794686,
      "loss": 0.4454,
      "step": 17900
    },
    {
      "epoch": 2.9866888519134775,
      "grad_norm": 0.4425393044948578,
      "learning_rate": 0.0004998056064423617,
      "loss": 0.4434,
      "step": 17950
    },
    {
      "epoch": 2.995008319467554,
      "grad_norm": 0.2697887718677521,
      "learning_rate": 0.000499805064805255,
      "loss": 0.4465,
      "step": 18000
    },
    {
      "epoch": 2.995008319467554,
      "eval_accuracy": 0.773348269518191,
      "eval_auc": 0.8564303811560599,
      "eval_f1": 0.7921942795235261,
      "eval_loss": 0.4447154104709625,
      "eval_precision": 0.7310888599612942,
      "eval_recall": 0.8644458620346581,
      "eval_runtime": 3118.3883,
      "eval_samples_per_second": 109.64,
      "eval_steps_per_second": 13.705,
      "step": 18000
    }
  ],
  "logging_steps": 50,
  "max_steps": 90150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.267314096665788e+18,
  "train_batch_size": 512,
  "trial_name": null,
  "trial_params": null
}