{
  "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": 5.7405501646885835,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.7567,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17380350828170776,
      "step": 5,
      "valid_targets_mean": 4028.6,
      "valid_targets_min": 1301
    },
    {
      "epoch": 0.064,
      "grad_norm": 4.118357314424626,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.7794,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16295763850212097,
      "step": 10,
      "valid_targets_mean": 3602.9,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.096,
      "grad_norm": 2.3711570697521784,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.6953,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1843426674604416,
      "step": 15,
      "valid_targets_mean": 4286.8,
      "valid_targets_min": 2447
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.9634533863414915,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.6589,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1682567298412323,
      "step": 20,
      "valid_targets_mean": 3729.9,
      "valid_targets_min": 834
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7952405580239124,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.6197,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12355852127075195,
      "step": 25,
      "valid_targets_mean": 2978.4,
      "valid_targets_min": 627
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.5503818444826258,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.6075,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1376994252204895,
      "step": 30,
      "valid_targets_mean": 3608.1,
      "valid_targets_min": 695
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.46586641631710746,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5811,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16773051023483276,
      "step": 35,
      "valid_targets_mean": 4446.0,
      "valid_targets_min": 1349
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.35607786106529044,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.5498,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10739520937204361,
      "step": 40,
      "valid_targets_mean": 4363.8,
      "valid_targets_min": 1156
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.3260991863479282,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.5237,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09335952997207642,
      "step": 45,
      "valid_targets_mean": 2915.9,
      "valid_targets_min": 505
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3159004340990336,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.5187,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.100542351603508,
      "step": 50,
      "valid_targets_mean": 3278.5,
      "valid_targets_min": 891
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.2984583420410525,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.5075,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12002827227115631,
      "step": 55,
      "valid_targets_mean": 4398.0,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.2743579374092805,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.4896,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1225425973534584,
      "step": 60,
      "valid_targets_mean": 4135.6,
      "valid_targets_min": 748
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.3222173078672134,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.5092,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11498141288757324,
      "step": 65,
      "valid_targets_mean": 3207.6,
      "valid_targets_min": 666
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.30981783757604325,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.4782,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11429333686828613,
      "step": 70,
      "valid_targets_mean": 3716.7,
      "valid_targets_min": 770
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.26579237645780096,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.4919,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1003502607345581,
      "step": 75,
      "valid_targets_mean": 3828.2,
      "valid_targets_min": 442
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.30914927492134714,
      "learning_rate": 4e-05,
      "loss": 0.4723,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10817784070968628,
      "step": 80,
      "valid_targets_mean": 3202.1,
      "valid_targets_min": 712
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.2779595243502161,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.4625,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10495799779891968,
      "step": 85,
      "valid_targets_mean": 3934.1,
      "valid_targets_min": 659
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.3164902882432122,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.4583,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10236912965774536,
      "step": 90,
      "valid_targets_mean": 3006.6,
      "valid_targets_min": 670
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.2692165670736237,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4651,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12215817719697952,
      "step": 95,
      "valid_targets_mean": 4568.6,
      "valid_targets_min": 974
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3136264928708605,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.4522,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11707773059606552,
      "step": 100,
      "valid_targets_mean": 3752.2,
      "valid_targets_min": 886
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.28913071361897935,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.4475,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.125753715634346,
      "step": 105,
      "valid_targets_mean": 4481.6,
      "valid_targets_min": 760
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.29683528551147204,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4461,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12000398337841034,
      "step": 110,
      "valid_targets_mean": 4761.2,
      "valid_targets_min": 1467
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.2925993861951088,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.4432,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12122927606105804,
      "step": 115,
      "valid_targets_mean": 3916.0,
      "valid_targets_min": 576
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.3013806179777486,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.4506,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12075666338205338,
      "step": 120,
      "valid_targets_mean": 4072.9,
      "valid_targets_min": 1069
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.28788610284242316,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4472,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09057885408401489,
      "step": 125,
      "valid_targets_mean": 4109.1,
      "valid_targets_min": 1019
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.32368991570988725,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.4518,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11525657773017883,
      "step": 130,
      "valid_targets_mean": 3226.8,
      "valid_targets_min": 687
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.27547237765892657,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.4548,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11187707632780075,
      "step": 135,
      "valid_targets_mean": 3624.2,
      "valid_targets_min": 496
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.289130190723095,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.4363,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09981520473957062,
      "step": 140,
      "valid_targets_mean": 3737.1,
      "valid_targets_min": 729
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.27773197672385863,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.4187,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10650136321783066,
      "step": 145,
      "valid_targets_mean": 4227.4,
      "valid_targets_min": 812
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.29439901160515425,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.4305,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1241697445511818,
      "step": 150,
      "valid_targets_mean": 4391.5,
      "valid_targets_min": 987
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.28181169657784816,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.4237,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10098722577095032,
      "step": 155,
      "valid_targets_mean": 4208.0,
      "valid_targets_min": 676
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.28019611015659524,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.431,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09703908860683441,
      "step": 160,
      "valid_targets_mean": 3175.7,
      "valid_targets_min": 575
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.30692174422902624,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.4317,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10665266215801239,
      "step": 165,
      "valid_targets_mean": 3968.7,
      "valid_targets_min": 683
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.2975111095131961,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.429,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1001289114356041,
      "step": 170,
      "valid_targets_mean": 3356.7,
      "valid_targets_min": 858
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.2841208133488064,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.4194,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10832270979881287,
      "step": 175,
      "valid_targets_mean": 3420.8,
      "valid_targets_min": 535
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.2980451869737519,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.4215,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.104207344353199,
      "step": 180,
      "valid_targets_mean": 3845.6,
      "valid_targets_min": 1237
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.2967764663137351,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.4164,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09934418648481369,
      "step": 185,
      "valid_targets_mean": 3445.5,
      "valid_targets_min": 826
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.2569214808604888,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.4213,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10533586144447327,
      "step": 190,
      "valid_targets_mean": 4363.2,
      "valid_targets_min": 735
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.27515236270381244,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.4156,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11472062766551971,
      "step": 195,
      "valid_targets_mean": 5027.0,
      "valid_targets_min": 2016
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.2901317104574311,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.421,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10855934023857117,
      "step": 200,
      "valid_targets_mean": 3643.5,
      "valid_targets_min": 824
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.2728964948648713,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.4152,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08453670889139175,
      "step": 205,
      "valid_targets_mean": 3758.7,
      "valid_targets_min": 712
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.2913756424634491,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.4312,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12182488292455673,
      "step": 210,
      "valid_targets_mean": 4696.9,
      "valid_targets_min": 1310
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.28692467134761634,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.4268,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.096201092004776,
      "step": 215,
      "valid_targets_mean": 3428.6,
      "valid_targets_min": 544
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.28168645927261154,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.4154,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10372606664896011,
      "step": 220,
      "valid_targets_mean": 4333.8,
      "valid_targets_min": 615
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.27895984471528773,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.4217,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11365018784999847,
      "step": 225,
      "valid_targets_mean": 3794.8,
      "valid_targets_min": 747
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.2546486429954818,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.4207,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11202923208475113,
      "step": 230,
      "valid_targets_mean": 4884.8,
      "valid_targets_min": 1951
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.2963808362591912,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.4161,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11676149815320969,
      "step": 235,
      "valid_targets_mean": 4393.4,
      "valid_targets_min": 946
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.29411982686108756,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.4152,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08890153467655182,
      "step": 240,
      "valid_targets_mean": 3607.9,
      "valid_targets_min": 627
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.29446107492001333,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.415,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10130105912685394,
      "step": 245,
      "valid_targets_mean": 4165.3,
      "valid_targets_min": 1380
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.297047538131773,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.4301,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10856887698173523,
      "step": 250,
      "valid_targets_mean": 4042.0,
      "valid_targets_min": 988
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.28428011285127336,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.4155,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08923467248678207,
      "step": 255,
      "valid_targets_mean": 4024.3,
      "valid_targets_min": 699
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.2741367953196807,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.41,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10443554818630219,
      "step": 260,
      "valid_targets_mean": 4403.8,
      "valid_targets_min": 840
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.27989290099156033,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.415,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09717078506946564,
      "step": 265,
      "valid_targets_mean": 3149.2,
      "valid_targets_min": 717
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.2657752304737816,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.4045,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11956058442592621,
      "step": 270,
      "valid_targets_mean": 4721.5,
      "valid_targets_min": 834
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.28588325118108937,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.4091,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11287073791027069,
      "step": 275,
      "valid_targets_mean": 3619.1,
      "valid_targets_min": 1147
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.26275468136089825,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.4108,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11017081886529922,
      "step": 280,
      "valid_targets_mean": 4387.8,
      "valid_targets_min": 1588
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.3166954873589347,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.4042,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11849084496498108,
      "step": 285,
      "valid_targets_mean": 4148.1,
      "valid_targets_min": 994
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.29787729159033155,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.4058,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09117233753204346,
      "step": 290,
      "valid_targets_mean": 4229.9,
      "valid_targets_min": 672
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.2700170652845831,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.4185,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10145064443349838,
      "step": 295,
      "valid_targets_mean": 3714.1,
      "valid_targets_min": 805
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.2719698667308842,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.4115,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12178967893123627,
      "step": 300,
      "valid_targets_mean": 4512.9,
      "valid_targets_min": 718
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.2853718609509193,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.4114,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09692707657814026,
      "step": 305,
      "valid_targets_mean": 3915.2,
      "valid_targets_min": 605
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.29811909104016,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.4222,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.090706467628479,
      "step": 310,
      "valid_targets_mean": 2961.4,
      "valid_targets_min": 663
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.2911674565723724,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.41,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10237012803554535,
      "step": 315,
      "valid_targets_mean": 3974.5,
      "valid_targets_min": 1309
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.29006617752098124,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.4129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1168978363275528,
      "step": 320,
      "valid_targets_mean": 4476.9,
      "valid_targets_min": 842
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.29713826669511567,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.4019,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11255374550819397,
      "step": 325,
      "valid_targets_mean": 4824.8,
      "valid_targets_min": 962
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.24446713000923223,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.3927,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08244487643241882,
      "step": 330,
      "valid_targets_mean": 4150.0,
      "valid_targets_min": 631
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.2979763044478798,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.3994,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08973608911037445,
      "step": 335,
      "valid_targets_mean": 3546.3,
      "valid_targets_min": 831
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.3111312990776006,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.4016,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12511217594146729,
      "step": 340,
      "valid_targets_mean": 4074.4,
      "valid_targets_min": 798
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.2716874405549092,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.4001,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08802570402622223,
      "step": 345,
      "valid_targets_mean": 4214.4,
      "valid_targets_min": 1315
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.268035602413646,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.3948,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10543382167816162,
      "step": 350,
      "valid_targets_mean": 4785.6,
      "valid_targets_min": 773
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.2725940756417308,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.3934,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09025853872299194,
      "step": 355,
      "valid_targets_mean": 3809.7,
      "valid_targets_min": 779
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.2884628727307354,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.3937,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09721674025058746,
      "step": 360,
      "valid_targets_mean": 3914.7,
      "valid_targets_min": 720
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.27759745229175836,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.4013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08376447856426239,
      "step": 365,
      "valid_targets_mean": 3270.2,
      "valid_targets_min": 1164
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.27066268352143985,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.4004,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10591837763786316,
      "step": 370,
      "valid_targets_mean": 4301.1,
      "valid_targets_min": 843
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.2611623318151007,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3997,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1060149222612381,
      "step": 375,
      "valid_targets_mean": 4606.2,
      "valid_targets_min": 475
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.2700087025996874,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.4004,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09458446502685547,
      "step": 380,
      "valid_targets_mean": 3768.1,
      "valid_targets_min": 724
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.2882146725134982,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.3962,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11932457983493805,
      "step": 385,
      "valid_targets_mean": 4442.6,
      "valid_targets_min": 1490
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.2704198360211367,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.3975,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08749280869960785,
      "step": 390,
      "valid_targets_mean": 3546.4,
      "valid_targets_min": 603
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.27489586292402585,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.3978,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10903258621692657,
      "step": 395,
      "valid_targets_mean": 4648.9,
      "valid_targets_min": 1146
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.25762159139742524,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.3947,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08013826608657837,
      "step": 400,
      "valid_targets_mean": 3459.9,
      "valid_targets_min": 484
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.29111493056100934,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.4058,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09682591259479523,
      "step": 405,
      "valid_targets_mean": 3791.9,
      "valid_targets_min": 589
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.2904262273724528,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.4028,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0949597954750061,
      "step": 410,
      "valid_targets_mean": 3705.6,
      "valid_targets_min": 861
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.2764867102294946,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.3977,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09528416395187378,
      "step": 415,
      "valid_targets_mean": 3434.7,
      "valid_targets_min": 615
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.2746210141743562,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.4036,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10299257189035416,
      "step": 420,
      "valid_targets_mean": 3960.3,
      "valid_targets_min": 644
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.2509818521978272,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.3979,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10765567421913147,
      "step": 425,
      "valid_targets_mean": 5254.7,
      "valid_targets_min": 920
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.2939579884871459,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.3976,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1211467981338501,
      "step": 430,
      "valid_targets_mean": 4082.0,
      "valid_targets_min": 585
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.27537761615413714,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.391,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09925626218318939,
      "step": 435,
      "valid_targets_mean": 3994.1,
      "valid_targets_min": 619
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.25402119708320314,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3926,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10358227789402008,
      "step": 440,
      "valid_targets_mean": 5312.6,
      "valid_targets_min": 997
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.26728419643998946,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.4027,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08086955547332764,
      "step": 445,
      "valid_targets_mean": 3165.5,
      "valid_targets_min": 972
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.2636021477366375,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3976,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08937904983758926,
      "step": 450,
      "valid_targets_mean": 3714.8,
      "valid_targets_min": 535
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.2818910117875375,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.3873,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10908740758895874,
      "step": 455,
      "valid_targets_mean": 3477.9,
      "valid_targets_min": 956
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.2681224683219548,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.3965,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10116074979305267,
      "step": 460,
      "valid_targets_mean": 4017.2,
      "valid_targets_min": 494
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.28753730381790465,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.3926,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09378722310066223,
      "step": 465,
      "valid_targets_mean": 4233.3,
      "valid_targets_min": 783
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.27095437923046495,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.3944,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10863704979419708,
      "step": 470,
      "valid_targets_mean": 3905.1,
      "valid_targets_min": 924
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.25356106765252273,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.3944,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10768838226795197,
      "step": 475,
      "valid_targets_mean": 5093.1,
      "valid_targets_min": 2449
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.2721619342684793,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.3951,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10738348960876465,
      "step": 480,
      "valid_targets_mean": 4335.3,
      "valid_targets_min": 597
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.25518732340483063,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.3832,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1026557981967926,
      "step": 485,
      "valid_targets_mean": 4713.4,
      "valid_targets_min": 839
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.26079119811850293,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.3882,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09954249858856201,
      "step": 490,
      "valid_targets_mean": 4218.7,
      "valid_targets_min": 560
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.2807324596643295,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.3931,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09099026024341583,
      "step": 495,
      "valid_targets_mean": 3611.8,
      "valid_targets_min": 926
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.26487858125661595,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.3919,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0979473888874054,
      "step": 500,
      "valid_targets_mean": 3818.9,
      "valid_targets_min": 761
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.2555104373970932,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3912,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10479137301445007,
      "step": 505,
      "valid_targets_mean": 4825.6,
      "valid_targets_min": 610
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.2759366870779462,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3857,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09441757202148438,
      "step": 510,
      "valid_targets_mean": 3856.2,
      "valid_targets_min": 580
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.2655260723692603,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.3884,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09593252837657928,
      "step": 515,
      "valid_targets_mean": 3887.6,
      "valid_targets_min": 724
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.2444339074369139,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3929,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10919459164142609,
      "step": 520,
      "valid_targets_mean": 5236.3,
      "valid_targets_min": 1160
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.28022445189928447,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.3775,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08786538243293762,
      "step": 525,
      "valid_targets_mean": 3148.2,
      "valid_targets_min": 617
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.2695890785094206,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3767,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10315999388694763,
      "step": 530,
      "valid_targets_mean": 4860.6,
      "valid_targets_min": 607
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.2688143154625659,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.3878,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10301923751831055,
      "step": 535,
      "valid_targets_mean": 4068.3,
      "valid_targets_min": 760
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.2485960474064065,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.3957,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09227894991636276,
      "step": 540,
      "valid_targets_mean": 4552.3,
      "valid_targets_min": 855
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.26406506542236524,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.3768,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08604507148265839,
      "step": 545,
      "valid_targets_mean": 3193.4,
      "valid_targets_min": 631
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.28601007183878213,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.3891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09989183396100998,
      "step": 550,
      "valid_targets_mean": 4043.6,
      "valid_targets_min": 1062
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.2593652235958659,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3918,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09305549412965775,
      "step": 555,
      "valid_targets_mean": 3571.2,
      "valid_targets_min": 787
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.2806112076961371,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.3973,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11401621997356415,
      "step": 560,
      "valid_targets_mean": 4356.2,
      "valid_targets_min": 544
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.26438550224931556,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.3853,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09649631381034851,
      "step": 565,
      "valid_targets_mean": 4069.6,
      "valid_targets_min": 762
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.25444260633686594,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3894,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09096725285053253,
      "step": 570,
      "valid_targets_mean": 3807.6,
      "valid_targets_min": 627
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.2906114935948601,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.3904,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12105219066143036,
      "step": 575,
      "valid_targets_mean": 4758.1,
      "valid_targets_min": 1551
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.2849797377992677,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.3925,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0914318785071373,
      "step": 580,
      "valid_targets_mean": 3668.1,
      "valid_targets_min": 965
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.26884762228448145,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.3873,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11145298182964325,
      "step": 585,
      "valid_targets_mean": 4173.0,
      "valid_targets_min": 748
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.23474625858418305,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3774,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08823410421609879,
      "step": 590,
      "valid_targets_mean": 4066.7,
      "valid_targets_min": 949
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.25981641188064936,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.3844,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08829885721206665,
      "step": 595,
      "valid_targets_mean": 3905.6,
      "valid_targets_min": 1145
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.24814170802758997,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.3893,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07710494101047516,
      "step": 600,
      "valid_targets_mean": 3528.4,
      "valid_targets_min": 673
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.26538616985869273,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3842,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08873239159584045,
      "step": 605,
      "valid_targets_mean": 3282.8,
      "valid_targets_min": 672
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.2534875874697522,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.3955,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10614732652902603,
      "step": 610,
      "valid_targets_mean": 4293.0,
      "valid_targets_min": 633
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.24258039429477687,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.385,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09797520190477371,
      "step": 615,
      "valid_targets_mean": 4737.0,
      "valid_targets_min": 841
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.26467341402053607,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.389,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1028120294213295,
      "step": 620,
      "valid_targets_mean": 3816.8,
      "valid_targets_min": 1621
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.2622407218102556,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.3848,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10032577812671661,
      "step": 625,
      "valid_targets_mean": 4414.3,
      "valid_targets_min": 996
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.2735226414796938,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3832,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10025861114263535,
      "step": 630,
      "valid_targets_mean": 3794.0,
      "valid_targets_min": 1098
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.2616831477515809,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.3891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09857222437858582,
      "step": 635,
      "valid_targets_mean": 3492.7,
      "valid_targets_min": 610
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.2564684895303431,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.3799,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10000475496053696,
      "step": 640,
      "valid_targets_mean": 4033.1,
      "valid_targets_min": 1075
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.24515128829129684,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.3813,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0976802408695221,
      "step": 645,
      "valid_targets_mean": 4291.7,
      "valid_targets_min": 679
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.2519331481433472,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.3793,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07941633462905884,
      "step": 650,
      "valid_targets_mean": 3458.9,
      "valid_targets_min": 524
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.2666390566983211,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3811,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10504255443811417,
      "step": 655,
      "valid_targets_mean": 4640.8,
      "valid_targets_min": 1832
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.2784965710848798,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.3872,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09621089696884155,
      "step": 660,
      "valid_targets_mean": 3483.3,
      "valid_targets_min": 900
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.26661749606819596,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.3861,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09312206506729126,
      "step": 665,
      "valid_targets_mean": 3506.9,
      "valid_targets_min": 689
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.2634770624974283,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3799,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08135552704334259,
      "step": 670,
      "valid_targets_mean": 3128.4,
      "valid_targets_min": 633
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.24437543241558615,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3678,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10483628511428833,
      "step": 675,
      "valid_targets_mean": 4867.8,
      "valid_targets_min": 814
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.2442667141849187,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3829,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09347662329673767,
      "step": 680,
      "valid_targets_mean": 4420.1,
      "valid_targets_min": 667
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.27056999224802303,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.3849,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08870618045330048,
      "step": 685,
      "valid_targets_mean": 3327.4,
      "valid_targets_min": 810
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.2596022356965633,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3873,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10243427753448486,
      "step": 690,
      "valid_targets_mean": 4198.4,
      "valid_targets_min": 562
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.254273144137506,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.3821,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11028210818767548,
      "step": 695,
      "valid_targets_mean": 4355.8,
      "valid_targets_min": 697
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.2753527264558269,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.3906,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08981969952583313,
      "step": 700,
      "valid_targets_mean": 3440.1,
      "valid_targets_min": 693
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.2580146245246029,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3793,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10600106418132782,
      "step": 705,
      "valid_targets_mean": 4351.8,
      "valid_targets_min": 989
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.2657866427649952,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.3853,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10807494819164276,
      "step": 710,
      "valid_targets_mean": 4170.8,
      "valid_targets_min": 832
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.25426681062291456,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.3871,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1082836389541626,
      "step": 715,
      "valid_targets_mean": 4136.6,
      "valid_targets_min": 960
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.23643011778074544,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3799,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08495372533798218,
      "step": 720,
      "valid_targets_mean": 3793.1,
      "valid_targets_min": 1182
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.26215860215692277,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.3891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11342902481555939,
      "step": 725,
      "valid_targets_mean": 4220.9,
      "valid_targets_min": 1024
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.24797440452234024,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.3775,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08319179713726044,
      "step": 730,
      "valid_targets_mean": 3530.2,
      "valid_targets_min": 627
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.2679324054330442,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.3855,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0951373502612114,
      "step": 735,
      "valid_targets_mean": 3834.3,
      "valid_targets_min": 1019
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.26023484147122594,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3838,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09338989108800888,
      "step": 740,
      "valid_targets_mean": 3761.6,
      "valid_targets_min": 719
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.247354010218026,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.3845,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09644794464111328,
      "step": 745,
      "valid_targets_mean": 4205.1,
      "valid_targets_min": 1813
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.24027256426265428,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.3757,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09103810042142868,
      "step": 750,
      "valid_targets_mean": 3961.3,
      "valid_targets_min": 747
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.24737679779763444,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.3807,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09094603359699249,
      "step": 755,
      "valid_targets_mean": 4646.4,
      "valid_targets_min": 580
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.2661167677882559,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3916,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08691633492708206,
      "step": 760,
      "valid_targets_mean": 3201.0,
      "valid_targets_min": 762
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.2379386229615173,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3829,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10131211578845978,
      "step": 765,
      "valid_targets_mean": 4248.2,
      "valid_targets_min": 1315
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.2684414620796076,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.3808,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10655556619167328,
      "step": 770,
      "valid_targets_mean": 3972.6,
      "valid_targets_min": 668
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.24538242079579548,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.383,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10726399719715118,
      "step": 775,
      "valid_targets_mean": 4565.1,
      "valid_targets_min": 1007
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.24317743955775065,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3951,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08870485424995422,
      "step": 780,
      "valid_targets_mean": 3571.8,
      "valid_targets_min": 580
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.478858896945483,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3805,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.36933785676956177,
      "step": 785,
      "valid_targets_mean": 4053.2,
      "valid_targets_min": 645
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.36933785676956177,
      "step": 785,
      "total_flos": 1.3498462023419167e+18,
      "train_loss": 0.4199095495187553,
      "train_runtime": 16751.3373,
      "train_samples_per_second": 2.984,
      "train_steps_per_second": 0.047,
      "valid_targets_mean": 4053.2,
      "valid_targets_min": 645
    }
  ],
  "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": 1.3498462023419167e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}