{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 785,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032,
      "grad_norm": 5.541497684638077,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.7201,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.20204728841781616,
      "step": 5,
      "valid_targets_mean": 3694.1,
      "valid_targets_min": 918
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.5800693611841905,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.6795,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1315670758485794,
      "step": 10,
      "valid_targets_mean": 4516.5,
      "valid_targets_min": 765
    },
    {
      "epoch": 0.096,
      "grad_norm": 1.9884248548900978,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.6312,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13017678260803223,
      "step": 15,
      "valid_targets_mean": 4166.5,
      "valid_targets_min": 845
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.9233035770274349,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.6036,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.18069788813591003,
      "step": 20,
      "valid_targets_mean": 5295.3,
      "valid_targets_min": 864
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.725540551324882,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.5716,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1643570065498352,
      "step": 25,
      "valid_targets_mean": 5433.7,
      "valid_targets_min": 1289
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.5509203398092252,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.5351,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11187565326690674,
      "step": 30,
      "valid_targets_mean": 2928.9,
      "valid_targets_min": 793
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.4718786929689295,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5297,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15530270338058472,
      "step": 35,
      "valid_targets_mean": 4836.8,
      "valid_targets_min": 881
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.4068295200643166,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.4873,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12239599227905273,
      "step": 40,
      "valid_targets_mean": 3672.7,
      "valid_targets_min": 1172
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.34726917547603475,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.4779,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12165165692567825,
      "step": 45,
      "valid_targets_mean": 4045.2,
      "valid_targets_min": 834
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.29930653523131256,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.4638,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10612066090106964,
      "step": 50,
      "valid_targets_mean": 2865.4,
      "valid_targets_min": 630
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.3534925450119451,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.4716,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12155529856681824,
      "step": 55,
      "valid_targets_mean": 3224.7,
      "valid_targets_min": 530
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.2698100119387162,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.4361,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07520334422588348,
      "step": 60,
      "valid_targets_mean": 3178.1,
      "valid_targets_min": 672
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.31765210809501726,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.4641,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12320488691329956,
      "step": 65,
      "valid_targets_mean": 4320.3,
      "valid_targets_min": 814
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.2981457203142397,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.457,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09969944506883621,
      "step": 70,
      "valid_targets_mean": 2920.1,
      "valid_targets_min": 701
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2521133577518038,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.4469,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10771138221025467,
      "step": 75,
      "valid_targets_mean": 4220.4,
      "valid_targets_min": 902
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.34779990596233723,
      "learning_rate": 4e-05,
      "loss": 0.4457,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.141748309135437,
      "step": 80,
      "valid_targets_mean": 3178.2,
      "valid_targets_min": 619
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.2666291404893624,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.431,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.112694151699543,
      "step": 85,
      "valid_targets_mean": 5044.8,
      "valid_targets_min": 1099
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.2824399829410414,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.4267,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10418764501810074,
      "step": 90,
      "valid_targets_mean": 3889.9,
      "valid_targets_min": 699
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.31908682280434025,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4239,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09118764102458954,
      "step": 95,
      "valid_targets_mean": 2867.9,
      "valid_targets_min": 850
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.2549712848477804,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.4125,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09206928312778473,
      "step": 100,
      "valid_targets_mean": 4052.6,
      "valid_targets_min": 491
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.45582965229062333,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.4312,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09463967382907867,
      "step": 105,
      "valid_targets_mean": 2923.5,
      "valid_targets_min": 830
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.2707218225881968,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4085,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09323331713676453,
      "step": 110,
      "valid_targets_mean": 4330.3,
      "valid_targets_min": 723
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.3153755815756895,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.4184,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1101507917046547,
      "step": 115,
      "valid_targets_mean": 4692.8,
      "valid_targets_min": 711
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.3295458677006816,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.408,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08722187578678131,
      "step": 120,
      "valid_targets_mean": 2657.8,
      "valid_targets_min": 813
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3038603723433472,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4321,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11620064079761505,
      "step": 125,
      "valid_targets_mean": 3377.4,
      "valid_targets_min": 634
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.28009890305331936,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.4124,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09834864735603333,
      "step": 130,
      "valid_targets_mean": 3717.2,
      "valid_targets_min": 695
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.2566202226762974,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.3932,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07297708839178085,
      "step": 135,
      "valid_targets_mean": 2894.3,
      "valid_targets_min": 463
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.31636429084131246,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.4012,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08311374485492706,
      "step": 140,
      "valid_targets_mean": 4278.8,
      "valid_targets_min": 812
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.2576702240508463,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.4093,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10160231590270996,
      "step": 145,
      "valid_targets_mean": 4623.7,
      "valid_targets_min": 662
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.2877741398086508,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.4071,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1042058914899826,
      "step": 150,
      "valid_targets_mean": 3411.9,
      "valid_targets_min": 596
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.2938652286141047,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.4119,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10854722559452057,
      "step": 155,
      "valid_targets_mean": 3161.6,
      "valid_targets_min": 555
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.33200094179793993,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.3916,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08427420258522034,
      "step": 160,
      "valid_targets_mean": 2848.1,
      "valid_targets_min": 719
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.26215404770192946,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.3837,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06924141198396683,
      "step": 165,
      "valid_targets_mean": 2998.8,
      "valid_targets_min": 1000
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.3016475809185609,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.3976,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07719582319259644,
      "step": 170,
      "valid_targets_mean": 2974.9,
      "valid_targets_min": 907
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.3047060279415802,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.3772,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0757255107164383,
      "step": 175,
      "valid_targets_mean": 2829.6,
      "valid_targets_min": 764
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.3273636057159624,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.3891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1268865019083023,
      "step": 180,
      "valid_targets_mean": 4603.5,
      "valid_targets_min": 723
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.3125871525191448,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.4073,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12131393700838089,
      "step": 185,
      "valid_targets_mean": 4125.5,
      "valid_targets_min": 639
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.32796667514693834,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.3852,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10205736756324768,
      "step": 190,
      "valid_targets_mean": 3710.7,
      "valid_targets_min": 878
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.24475735520158587,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.3728,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10854168981313705,
      "step": 195,
      "valid_targets_mean": 6517.8,
      "valid_targets_min": 837
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.8583151313832471,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.3954,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07846558094024658,
      "step": 200,
      "valid_targets_mean": 2350.1,
      "valid_targets_min": 732
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.30922015694302607,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.4072,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08257310092449188,
      "step": 205,
      "valid_targets_mean": 3387.9,
      "valid_targets_min": 864
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.25189684546137714,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.3993,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09581450372934341,
      "step": 210,
      "valid_targets_mean": 3629.7,
      "valid_targets_min": 731
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.29112659892347204,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.3882,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06733721494674683,
      "step": 215,
      "valid_targets_mean": 2276.2,
      "valid_targets_min": 667
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.30010766829015695,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.4001,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1396431028842926,
      "step": 220,
      "valid_targets_mean": 4677.5,
      "valid_targets_min": 672
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.33649560235455167,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.3982,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07678442448377609,
      "step": 225,
      "valid_targets_mean": 2519.9,
      "valid_targets_min": 898
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.2987276715695913,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.3941,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10258451104164124,
      "step": 230,
      "valid_targets_mean": 3978.2,
      "valid_targets_min": 762
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.26735894969723445,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.3856,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09502792358398438,
      "step": 235,
      "valid_targets_mean": 5169.4,
      "valid_targets_min": 662
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.2790211030720928,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.3793,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08720822632312775,
      "step": 240,
      "valid_targets_mean": 3078.2,
      "valid_targets_min": 1129
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.28420101467873393,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.3923,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06919790804386139,
      "step": 245,
      "valid_targets_mean": 2619.1,
      "valid_targets_min": 686
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.28761286823955284,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.3985,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09496282041072845,
      "step": 250,
      "valid_targets_mean": 5359.2,
      "valid_targets_min": 701
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.28798208612082316,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.3753,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1058167815208435,
      "step": 255,
      "valid_targets_mean": 4252.7,
      "valid_targets_min": 869
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.2905843016289666,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.3973,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07735498249530792,
      "step": 260,
      "valid_targets_mean": 3003.8,
      "valid_targets_min": 528
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.2399842469644208,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.3856,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07514525949954987,
      "step": 265,
      "valid_targets_mean": 4324.5,
      "valid_targets_min": 711
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.2769708989910312,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.3741,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08158373832702637,
      "step": 270,
      "valid_targets_mean": 3444.5,
      "valid_targets_min": 832
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.2865165779666472,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.3742,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10593311488628387,
      "step": 275,
      "valid_targets_mean": 4072.2,
      "valid_targets_min": 753
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.25372481435497135,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.3813,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10056015104055405,
      "step": 280,
      "valid_targets_mean": 4768.8,
      "valid_targets_min": 1129
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.2678766717461159,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.3697,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0960194319486618,
      "step": 285,
      "valid_targets_mean": 4018.4,
      "valid_targets_min": 684
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.26041593540773744,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.3824,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10017910599708557,
      "step": 290,
      "valid_targets_mean": 4154.5,
      "valid_targets_min": 636
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.26247724384317117,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.3992,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07477947324514389,
      "step": 295,
      "valid_targets_mean": 3326.6,
      "valid_targets_min": 815
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.2498670350702193,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.3685,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09179487824440002,
      "step": 300,
      "valid_targets_mean": 4956.8,
      "valid_targets_min": 819
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.2369503923852665,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.3797,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09301997721195221,
      "step": 305,
      "valid_targets_mean": 5283.4,
      "valid_targets_min": 711
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.28546224115455,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.379,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1093452051281929,
      "step": 310,
      "valid_targets_mean": 5712.8,
      "valid_targets_min": 623
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.2957261573831895,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.3803,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.073640838265419,
      "step": 315,
      "valid_targets_mean": 2807.1,
      "valid_targets_min": 676
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.2797029601720705,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.3644,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0723484680056572,
      "step": 320,
      "valid_targets_mean": 4557.2,
      "valid_targets_min": 626
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.25397901615246865,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.3755,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09201750159263611,
      "step": 325,
      "valid_targets_mean": 4751.2,
      "valid_targets_min": 800
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.2755393810570339,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.3752,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09428532421588898,
      "step": 330,
      "valid_targets_mean": 4889.7,
      "valid_targets_min": 817
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.32318560777114086,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.3801,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14423143863677979,
      "step": 335,
      "valid_targets_mean": 3977.7,
      "valid_targets_min": 613
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.24158332518684553,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.3655,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05750841647386551,
      "step": 340,
      "valid_targets_mean": 2743.6,
      "valid_targets_min": 767
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.2828544403010422,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.3644,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09143859148025513,
      "step": 345,
      "valid_targets_mean": 3692.8,
      "valid_targets_min": 679
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.30087857652855765,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.3843,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10722360759973526,
      "step": 350,
      "valid_targets_mean": 4067.9,
      "valid_targets_min": 749
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.27604332089737327,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.3884,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09173067659139633,
      "step": 355,
      "valid_targets_mean": 4361.7,
      "valid_targets_min": 761
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.41830966168714057,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.3519,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08942142128944397,
      "step": 360,
      "valid_targets_mean": 4370.5,
      "valid_targets_min": 955
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.26532006086060894,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.3687,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08756840229034424,
      "step": 365,
      "valid_targets_mean": 3483.9,
      "valid_targets_min": 709
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.25399461718436944,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.3553,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06391710042953491,
      "step": 370,
      "valid_targets_mean": 3258.9,
      "valid_targets_min": 762
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.26459151394063957,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3797,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09428556263446808,
      "step": 375,
      "valid_targets_mean": 3723.7,
      "valid_targets_min": 554
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.2592657995166384,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.367,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10876058042049408,
      "step": 380,
      "valid_targets_mean": 5299.8,
      "valid_targets_min": 820
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.3133546870917826,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.3883,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09236639738082886,
      "step": 385,
      "valid_targets_mean": 3117.1,
      "valid_targets_min": 682
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.2805498780921623,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.3518,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11789987981319427,
      "step": 390,
      "valid_targets_mean": 5694.6,
      "valid_targets_min": 657
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.2468211553990243,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.3637,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07164323329925537,
      "step": 395,
      "valid_targets_mean": 3688.2,
      "valid_targets_min": 734
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.25507421149574355,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.3798,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08426293730735779,
      "step": 400,
      "valid_targets_mean": 4809.1,
      "valid_targets_min": 844
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.25872300499927087,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.3658,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09084966778755188,
      "step": 405,
      "valid_targets_mean": 3788.1,
      "valid_targets_min": 596
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.24955127728931853,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.3499,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07429323345422745,
      "step": 410,
      "valid_targets_mean": 4410.4,
      "valid_targets_min": 677
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.28017640413377953,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.369,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0956093892455101,
      "step": 415,
      "valid_targets_mean": 4106.1,
      "valid_targets_min": 1154
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.2532898705270473,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.3711,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1108713299036026,
      "step": 420,
      "valid_targets_mean": 5110.1,
      "valid_targets_min": 949
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.2730879181166748,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.3695,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1020272970199585,
      "step": 425,
      "valid_targets_mean": 4568.8,
      "valid_targets_min": 758
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.2832238642802859,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.3702,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08114699274301529,
      "step": 430,
      "valid_targets_mean": 2751.1,
      "valid_targets_min": 805
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.29396256819439887,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.3769,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08736386895179749,
      "step": 435,
      "valid_targets_mean": 4836.0,
      "valid_targets_min": 824
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.293725631024192,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3629,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05763329192996025,
      "step": 440,
      "valid_targets_mean": 2313.6,
      "valid_targets_min": 765
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.2877941893706849,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.3755,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09069660305976868,
      "step": 445,
      "valid_targets_mean": 3941.3,
      "valid_targets_min": 739
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.29559513503682666,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3465,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08671572804450989,
      "step": 450,
      "valid_targets_mean": 4384.7,
      "valid_targets_min": 869
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.2926583293018059,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.3628,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07952509820461273,
      "step": 455,
      "valid_targets_mean": 4060.8,
      "valid_targets_min": 716
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.27213293831765606,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.3602,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09436945617198944,
      "step": 460,
      "valid_targets_mean": 3749.9,
      "valid_targets_min": 602
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.26583061680884507,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.3725,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0769558995962143,
      "step": 465,
      "valid_targets_mean": 3624.6,
      "valid_targets_min": 557
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.28691255129637394,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.3695,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0934470146894455,
      "step": 470,
      "valid_targets_mean": 3964.6,
      "valid_targets_min": 946
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.27694974177269244,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.369,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07107527554035187,
      "step": 475,
      "valid_targets_mean": 3270.2,
      "valid_targets_min": 705
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.2735034341978741,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.3475,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0945439338684082,
      "step": 480,
      "valid_targets_mean": 4498.7,
      "valid_targets_min": 275
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.23880890297429838,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.3451,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0911998599767685,
      "step": 485,
      "valid_targets_mean": 4708.6,
      "valid_targets_min": 739
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.27371534202194003,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.3649,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09913412481546402,
      "step": 490,
      "valid_targets_mean": 4566.2,
      "valid_targets_min": 661
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.27627616833465707,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.3702,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09827420115470886,
      "step": 495,
      "valid_targets_mean": 4863.6,
      "valid_targets_min": 780
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.30942075492779997,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.3541,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1027892604470253,
      "step": 500,
      "valid_targets_mean": 3456.2,
      "valid_targets_min": 737
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.3011990578205873,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3848,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10471899807453156,
      "step": 505,
      "valid_targets_mean": 3511.1,
      "valid_targets_min": 1126
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.25586343923489,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3531,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06394007802009583,
      "step": 510,
      "valid_targets_mean": 3619.6,
      "valid_targets_min": 801
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.247916108494188,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.3519,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06505729258060455,
      "step": 515,
      "valid_targets_mean": 3381.6,
      "valid_targets_min": 870
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.28193646168227443,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3641,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07794380187988281,
      "step": 520,
      "valid_targets_mean": 3805.7,
      "valid_targets_min": 808
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.25690023433577935,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.3456,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0881698876619339,
      "step": 525,
      "valid_targets_mean": 4807.1,
      "valid_targets_min": 665
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.2549757887509029,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3581,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07912416756153107,
      "step": 530,
      "valid_targets_mean": 3860.9,
      "valid_targets_min": 596
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.28204253021652476,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.3515,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09103786200284958,
      "step": 535,
      "valid_targets_mean": 3632.4,
      "valid_targets_min": 833
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.2943012232723084,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.3502,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10494515299797058,
      "step": 540,
      "valid_targets_mean": 4191.4,
      "valid_targets_min": 760
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.2601962964616993,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.361,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11226920038461685,
      "step": 545,
      "valid_targets_mean": 5768.5,
      "valid_targets_min": 1032
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.2771216636065589,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.3528,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0983453094959259,
      "step": 550,
      "valid_targets_mean": 4185.4,
      "valid_targets_min": 1072
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.29687442352238086,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3695,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10656087100505829,
      "step": 555,
      "valid_targets_mean": 3893.4,
      "valid_targets_min": 586
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.3258078619224109,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.3536,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09270378947257996,
      "step": 560,
      "valid_targets_mean": 3400.8,
      "valid_targets_min": 671
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.25100309171961954,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.3666,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09881480038166046,
      "step": 565,
      "valid_targets_mean": 5235.2,
      "valid_targets_min": 879
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.28813454432354996,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3666,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06468958407640457,
      "step": 570,
      "valid_targets_mean": 3096.8,
      "valid_targets_min": 672
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.2715471326270883,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.3537,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12461410462856293,
      "step": 575,
      "valid_targets_mean": 4765.1,
      "valid_targets_min": 1146
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.24995429633219066,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.3741,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08743943274021149,
      "step": 580,
      "valid_targets_mean": 3921.1,
      "valid_targets_min": 1066
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.2710951610884447,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.3494,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10593526065349579,
      "step": 585,
      "valid_targets_mean": 4516.6,
      "valid_targets_min": 745
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.24554789964053841,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3526,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07037960737943649,
      "step": 590,
      "valid_targets_mean": 4179.7,
      "valid_targets_min": 471
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.2740035607007115,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.3569,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11404471844434738,
      "step": 595,
      "valid_targets_mean": 4479.6,
      "valid_targets_min": 422
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.27857174556705694,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.3638,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11332307755947113,
      "step": 600,
      "valid_targets_mean": 4525.1,
      "valid_targets_min": 1350
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.27036674346841433,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3624,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09035474061965942,
      "step": 605,
      "valid_targets_mean": 3910.2,
      "valid_targets_min": 789
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.26713392574410405,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.3476,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07560218870639801,
      "step": 610,
      "valid_targets_mean": 3616.6,
      "valid_targets_min": 685
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.2551820724608469,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.348,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06616517901420593,
      "step": 615,
      "valid_targets_mean": 2984.4,
      "valid_targets_min": 545
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.2805543365431347,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.3688,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09327379614114761,
      "step": 620,
      "valid_targets_mean": 3519.0,
      "valid_targets_min": 613
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.26253382838902783,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.3563,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08990518748760223,
      "step": 625,
      "valid_targets_mean": 4271.6,
      "valid_targets_min": 957
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.2752324527328725,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3569,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09592108428478241,
      "step": 630,
      "valid_targets_mean": 4334.0,
      "valid_targets_min": 825
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.28556223681750137,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.3666,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08014878630638123,
      "step": 635,
      "valid_targets_mean": 3147.9,
      "valid_targets_min": 371
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.2500322876350698,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.344,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09790447354316711,
      "step": 640,
      "valid_targets_mean": 4746.9,
      "valid_targets_min": 831
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.2658128147848985,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.3557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11421743780374527,
      "step": 645,
      "valid_targets_mean": 5214.5,
      "valid_targets_min": 760
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.2620287255007106,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.3577,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06395424902439117,
      "step": 650,
      "valid_targets_mean": 3036.3,
      "valid_targets_min": 919
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.27025685574336344,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3468,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0907667949795723,
      "step": 655,
      "valid_targets_mean": 4400.2,
      "valid_targets_min": 430
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.25882052653428583,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.3654,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09003841876983643,
      "step": 660,
      "valid_targets_mean": 4704.8,
      "valid_targets_min": 877
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.30004947633921647,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.3671,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06004023551940918,
      "step": 665,
      "valid_targets_mean": 2139.1,
      "valid_targets_min": 709
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.26732020980047305,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3547,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09256140887737274,
      "step": 670,
      "valid_targets_mean": 4849.9,
      "valid_targets_min": 958
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.2590174884621007,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3507,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07109656929969788,
      "step": 675,
      "valid_targets_mean": 3647.4,
      "valid_targets_min": 713
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.2753475580823456,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3456,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09951435029506683,
      "step": 680,
      "valid_targets_mean": 4019.5,
      "valid_targets_min": 738
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.25668120929058674,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.3508,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08455836772918701,
      "step": 685,
      "valid_targets_mean": 3837.6,
      "valid_targets_min": 916
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.7219968858898081,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3436,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08186328411102295,
      "step": 690,
      "valid_targets_mean": 2692.8,
      "valid_targets_min": 830
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.2646298047140867,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.3422,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06586353480815887,
      "step": 695,
      "valid_targets_mean": 3125.6,
      "valid_targets_min": 779
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.27486921576830325,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.3557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10247328877449036,
      "step": 700,
      "valid_targets_mean": 4226.4,
      "valid_targets_min": 872
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.2521880662478794,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3593,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08326166868209839,
      "step": 705,
      "valid_targets_mean": 3732.3,
      "valid_targets_min": 745
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.27665376943053505,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.3519,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11586222052574158,
      "step": 710,
      "valid_targets_mean": 5429.9,
      "valid_targets_min": 659
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.2873076430878492,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.3577,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1280900090932846,
      "step": 715,
      "valid_targets_mean": 4937.1,
      "valid_targets_min": 1150
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.2613753606147518,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3599,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0687340646982193,
      "step": 720,
      "valid_targets_mean": 3370.6,
      "valid_targets_min": 715
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.23830838005131266,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.3432,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09013132750988007,
      "step": 725,
      "valid_targets_mean": 4541.8,
      "valid_targets_min": 954
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.26521568774090987,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.348,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05397915840148926,
      "step": 730,
      "valid_targets_mean": 2744.1,
      "valid_targets_min": 523
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.36669589815503967,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.3532,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08455473929643631,
      "step": 735,
      "valid_targets_mean": 3970.4,
      "valid_targets_min": 463
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.2800304158375605,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3537,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.055731166154146194,
      "step": 740,
      "valid_targets_mean": 2201.8,
      "valid_targets_min": 653
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.28968831896780245,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.358,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10389368236064911,
      "step": 745,
      "valid_targets_mean": 5040.3,
      "valid_targets_min": 676
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.23644895830233334,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.3384,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07365654408931732,
      "step": 750,
      "valid_targets_mean": 4284.6,
      "valid_targets_min": 679
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.2522036009533431,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.3493,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06476405262947083,
      "step": 755,
      "valid_targets_mean": 3923.4,
      "valid_targets_min": 742
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.2538354955682196,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3465,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08009777963161469,
      "step": 760,
      "valid_targets_mean": 3415.3,
      "valid_targets_min": 784
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.26003194613743286,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3677,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0702776163816452,
      "step": 765,
      "valid_targets_mean": 3007.7,
      "valid_targets_min": 957
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.2798975287738127,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.3516,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10448385775089264,
      "step": 770,
      "valid_targets_mean": 3724.1,
      "valid_targets_min": 898
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.22457486103536708,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.339,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06149273365736008,
      "step": 775,
      "valid_targets_mean": 3175.9,
      "valid_targets_min": 626
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.2530148403115037,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3584,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07958637177944183,
      "step": 780,
      "valid_targets_mean": 4460.8,
      "valid_targets_min": 615
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5731229517630574,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3488,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.34793925285339355,
      "step": 785,
      "valid_targets_mean": 3301.4,
      "valid_targets_min": 867
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.34793925285339355,
      "step": 785,
      "total_flos": 1.0929687562016522e+18,
      "train_loss": 0.3876832940775877,
      "train_runtime": 17345.5697,
      "train_samples_per_second": 2.883,
      "train_steps_per_second": 0.045,
      "valid_targets_mean": 3301.4,
      "valid_targets_min": 867
    }
  ],
  "logging_steps": 5,
  "max_steps": 785,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0929687562016522e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}