{
  "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": 6.392433372509941,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.7863,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.20333202183246613,
      "step": 5,
      "valid_targets_mean": 3878.7,
      "valid_targets_min": 581
    },
    {
      "epoch": 0.064,
      "grad_norm": 4.013784763817711,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.7335,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17291371524333954,
      "step": 10,
      "valid_targets_mean": 3427.0,
      "valid_targets_min": 612
    },
    {
      "epoch": 0.096,
      "grad_norm": 2.3168943802638826,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.6953,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16444209218025208,
      "step": 15,
      "valid_targets_mean": 3645.4,
      "valid_targets_min": 655
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.9259594841767825,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.6632,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14370016753673553,
      "step": 20,
      "valid_targets_mean": 3485.6,
      "valid_targets_min": 506
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.779730960970606,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.6308,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13291092216968536,
      "step": 25,
      "valid_targets_mean": 3412.2,
      "valid_targets_min": 677
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.5152238075058901,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.5981,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13552741706371307,
      "step": 30,
      "valid_targets_mean": 4053.1,
      "valid_targets_min": 696
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.4702302842481759,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5851,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15689809620380402,
      "step": 35,
      "valid_targets_mean": 4178.0,
      "valid_targets_min": 720
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.39573290647845955,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.5585,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14028018712997437,
      "step": 40,
      "valid_targets_mean": 3323.5,
      "valid_targets_min": 516
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.34993534068469734,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.5369,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1510724574327469,
      "step": 45,
      "valid_targets_mean": 5090.8,
      "valid_targets_min": 695
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.2965706274000795,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.5122,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11167290806770325,
      "step": 50,
      "valid_targets_mean": 4483.9,
      "valid_targets_min": 1121
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.30596422897723163,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.5028,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1316324919462204,
      "step": 55,
      "valid_targets_mean": 4031.8,
      "valid_targets_min": 522
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.2818197191413443,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.498,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10013847053050995,
      "step": 60,
      "valid_targets_mean": 3522.8,
      "valid_targets_min": 645
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.27954542599331306,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.4862,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14188680052757263,
      "step": 65,
      "valid_targets_mean": 5032.1,
      "valid_targets_min": 511
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.2987745130604576,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.482,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10456384718418121,
      "step": 70,
      "valid_targets_mean": 3347.8,
      "valid_targets_min": 870
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2830394327583852,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.4814,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11446298658847809,
      "step": 75,
      "valid_targets_mean": 3576.2,
      "valid_targets_min": 1193
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.2557980550377446,
      "learning_rate": 4e-05,
      "loss": 0.459,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10576022416353226,
      "step": 80,
      "valid_targets_mean": 4458.1,
      "valid_targets_min": 719
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.28579929513788704,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.4536,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10776900500059128,
      "step": 85,
      "valid_targets_mean": 3704.6,
      "valid_targets_min": 559
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.2889221443153764,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.4563,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13737119734287262,
      "step": 90,
      "valid_targets_mean": 4227.1,
      "valid_targets_min": 722
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.2717632358435459,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4452,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0985407754778862,
      "step": 95,
      "valid_targets_mean": 4334.2,
      "valid_targets_min": 661
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.27065033560334134,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.4564,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11091828346252441,
      "step": 100,
      "valid_targets_mean": 3830.3,
      "valid_targets_min": 627
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.27297941913826906,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.4573,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14249369502067566,
      "step": 105,
      "valid_targets_mean": 6018.1,
      "valid_targets_min": 771
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.2897034067562241,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4402,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1337655782699585,
      "step": 110,
      "valid_targets_mean": 4526.0,
      "valid_targets_min": 1520
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.2587803810250689,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.4332,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10224001109600067,
      "step": 115,
      "valid_targets_mean": 4012.1,
      "valid_targets_min": 691
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.29016288524164674,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.4555,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11725439876317978,
      "step": 120,
      "valid_targets_mean": 4485.7,
      "valid_targets_min": 716
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.36092903154063805,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4349,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10319697111845016,
      "step": 125,
      "valid_targets_mean": 4095.9,
      "valid_targets_min": 911
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.27905492037265434,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.4394,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10171744972467422,
      "step": 130,
      "valid_targets_mean": 3330.2,
      "valid_targets_min": 680
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.32376967770220905,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.4364,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12379811704158783,
      "step": 135,
      "valid_targets_mean": 4361.9,
      "valid_targets_min": 2162
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.28786943409541155,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.4522,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1159370020031929,
      "step": 140,
      "valid_targets_mean": 3891.2,
      "valid_targets_min": 741
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.2951485666316784,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.4438,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11070635914802551,
      "step": 145,
      "valid_targets_mean": 3564.1,
      "valid_targets_min": 708
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.31848274948306876,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.4423,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12361764907836914,
      "step": 150,
      "valid_targets_mean": 4296.2,
      "valid_targets_min": 1432
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.33243636091491874,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.4404,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11886661499738693,
      "step": 155,
      "valid_targets_mean": 4019.6,
      "valid_targets_min": 807
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.30567656300497476,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.4325,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11932887881994247,
      "step": 160,
      "valid_targets_mean": 4070.8,
      "valid_targets_min": 643
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.2757509485616067,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.4171,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0983552560210228,
      "step": 165,
      "valid_targets_mean": 4333.4,
      "valid_targets_min": 607
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.2899085096489028,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.4249,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11976086348295212,
      "step": 170,
      "valid_targets_mean": 4907.0,
      "valid_targets_min": 1556
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.27332187557589793,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.4246,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11746491491794586,
      "step": 175,
      "valid_targets_mean": 4552.8,
      "valid_targets_min": 1011
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.2652790818894624,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.4143,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12047027051448822,
      "step": 180,
      "valid_targets_mean": 4974.2,
      "valid_targets_min": 2923
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.2715122254756711,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.414,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11439437419176102,
      "step": 185,
      "valid_targets_mean": 4361.8,
      "valid_targets_min": 1157
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.3228931374283769,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.4316,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1057361364364624,
      "step": 190,
      "valid_targets_mean": 3297.6,
      "valid_targets_min": 742
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.2871794121630436,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.4338,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12501230835914612,
      "step": 195,
      "valid_targets_mean": 4784.6,
      "valid_targets_min": 821
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.29259965466267757,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.4183,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1052687019109726,
      "step": 200,
      "valid_targets_mean": 4254.4,
      "valid_targets_min": 930
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.28378460478191286,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.4268,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10018887370824814,
      "step": 205,
      "valid_targets_mean": 3303.4,
      "valid_targets_min": 817
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.28769132625474675,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.4134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09186606109142303,
      "step": 210,
      "valid_targets_mean": 3410.4,
      "valid_targets_min": 976
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.30246867963518775,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.4151,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12148073315620422,
      "step": 215,
      "valid_targets_mean": 4317.2,
      "valid_targets_min": 598
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.3266795625591887,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.4245,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11684419959783554,
      "step": 220,
      "valid_targets_mean": 4882.0,
      "valid_targets_min": 601
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.30436915531903647,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.417,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09582455456256866,
      "step": 225,
      "valid_targets_mean": 3790.4,
      "valid_targets_min": 595
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.26894452632641147,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.4234,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11669404804706573,
      "step": 230,
      "valid_targets_mean": 4575.8,
      "valid_targets_min": 726
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.2826506834617795,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.4209,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11476343870162964,
      "step": 235,
      "valid_targets_mean": 4214.8,
      "valid_targets_min": 767
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.2998125115812242,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.4113,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13182711601257324,
      "step": 240,
      "valid_targets_mean": 4447.1,
      "valid_targets_min": 1029
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.2650447923863934,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.4197,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09707813709974289,
      "step": 245,
      "valid_targets_mean": 3921.4,
      "valid_targets_min": 741
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.24986244106032818,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.4059,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11160187423229218,
      "step": 250,
      "valid_targets_mean": 5246.0,
      "valid_targets_min": 740
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.29232624600993573,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.4138,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09967529773712158,
      "step": 255,
      "valid_targets_mean": 3789.1,
      "valid_targets_min": 950
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.26098488748509174,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.408,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09568874537944794,
      "step": 260,
      "valid_targets_mean": 3921.2,
      "valid_targets_min": 758
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.28973769692698126,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.4156,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09107379615306854,
      "step": 265,
      "valid_targets_mean": 3219.7,
      "valid_targets_min": 946
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.2844976799567937,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.4098,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09754543751478195,
      "step": 270,
      "valid_targets_mean": 3920.6,
      "valid_targets_min": 752
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.26895164073391153,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.4125,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09589682519435883,
      "step": 275,
      "valid_targets_mean": 3786.6,
      "valid_targets_min": 1105
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.673713875249105,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.4152,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11353392899036407,
      "step": 280,
      "valid_targets_mean": 3856.1,
      "valid_targets_min": 862
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.3262149135411268,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.4111,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08625467121601105,
      "step": 285,
      "valid_targets_mean": 2924.9,
      "valid_targets_min": 642
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.28579420907876224,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.413,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08817364275455475,
      "step": 290,
      "valid_targets_mean": 3691.3,
      "valid_targets_min": 711
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.26322923306995594,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.4034,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09266021847724915,
      "step": 295,
      "valid_targets_mean": 4039.8,
      "valid_targets_min": 675
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.2733139550585241,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.4138,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.097264364361763,
      "step": 300,
      "valid_targets_mean": 3765.4,
      "valid_targets_min": 645
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.3089312402768751,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.4154,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10797446221113205,
      "step": 305,
      "valid_targets_mean": 3720.9,
      "valid_targets_min": 771
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.26788720481719935,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.3993,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10950686037540436,
      "step": 310,
      "valid_targets_mean": 4088.0,
      "valid_targets_min": 1049
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.3946782310914107,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.404,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09564638137817383,
      "step": 315,
      "valid_targets_mean": 3705.4,
      "valid_targets_min": 655
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.33172314938714054,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.4079,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12492215633392334,
      "step": 320,
      "valid_targets_mean": 3916.0,
      "valid_targets_min": 774
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.27370751165514795,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.3952,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10002350807189941,
      "step": 325,
      "valid_targets_mean": 3678.6,
      "valid_targets_min": 601
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.276204360301079,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.3982,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09459051489830017,
      "step": 330,
      "valid_targets_mean": 4414.2,
      "valid_targets_min": 719
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.26961372036662723,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.4019,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09708517789840698,
      "step": 335,
      "valid_targets_mean": 3878.8,
      "valid_targets_min": 668
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.2909876796289189,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.4018,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10152760148048401,
      "step": 340,
      "valid_targets_mean": 4209.8,
      "valid_targets_min": 678
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.2579041848615465,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.4007,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11158838123083115,
      "step": 345,
      "valid_targets_mean": 4669.5,
      "valid_targets_min": 790
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.3201490547233964,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.4098,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12716351449489594,
      "step": 350,
      "valid_targets_mean": 4234.3,
      "valid_targets_min": 608
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.29416498020012505,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.4034,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09254752844572067,
      "step": 355,
      "valid_targets_mean": 3754.1,
      "valid_targets_min": 483
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.2940327701345315,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.3983,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1163373589515686,
      "step": 360,
      "valid_targets_mean": 4291.6,
      "valid_targets_min": 885
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.2708021403047465,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.4082,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10339049994945526,
      "step": 365,
      "valid_targets_mean": 4598.2,
      "valid_targets_min": 807
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.26824622777428886,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.4013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08657452464103699,
      "step": 370,
      "valid_targets_mean": 3972.3,
      "valid_targets_min": 619
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.2823603784418457,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3923,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09838621318340302,
      "step": 375,
      "valid_targets_mean": 3567.2,
      "valid_targets_min": 549
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.2855326023347284,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.3887,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07984013855457306,
      "step": 380,
      "valid_targets_mean": 3190.5,
      "valid_targets_min": 868
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.2635608949404851,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.3972,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.106719471514225,
      "step": 385,
      "valid_targets_mean": 4622.6,
      "valid_targets_min": 1398
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.29332447615791285,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.4008,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10881468653678894,
      "step": 390,
      "valid_targets_mean": 4344.8,
      "valid_targets_min": 702
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.2994542523698844,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.3856,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09316129237413406,
      "step": 395,
      "valid_targets_mean": 3787.8,
      "valid_targets_min": 702
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.2750793613068623,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.3893,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10363581776618958,
      "step": 400,
      "valid_targets_mean": 3606.7,
      "valid_targets_min": 741
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.37133848874367914,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.3957,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11009484529495239,
      "step": 405,
      "valid_targets_mean": 4556.3,
      "valid_targets_min": 485
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.2590542056233833,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.3859,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10512518882751465,
      "step": 410,
      "valid_targets_mean": 5560.6,
      "valid_targets_min": 550
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.29307112484427317,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.3919,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08204679191112518,
      "step": 415,
      "valid_targets_mean": 2997.4,
      "valid_targets_min": 483
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.26682296153102414,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.3993,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07343417406082153,
      "step": 420,
      "valid_targets_mean": 3112.1,
      "valid_targets_min": 816
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.28603683907982996,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.3968,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10481780022382736,
      "step": 425,
      "valid_targets_mean": 3876.9,
      "valid_targets_min": 653
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.258632381058146,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.4005,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11329646408557892,
      "step": 430,
      "valid_targets_mean": 5110.6,
      "valid_targets_min": 1182
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.26842238376339367,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.3935,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09272165596485138,
      "step": 435,
      "valid_targets_mean": 3729.8,
      "valid_targets_min": 1037
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.27232586803317715,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3871,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11061014980077744,
      "step": 440,
      "valid_targets_mean": 4692.2,
      "valid_targets_min": 1102
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.2685990516821801,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.3964,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10506574809551239,
      "step": 445,
      "valid_targets_mean": 4644.6,
      "valid_targets_min": 1732
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.2796394266740902,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3911,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12312667071819305,
      "step": 450,
      "valid_targets_mean": 4376.6,
      "valid_targets_min": 1004
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.2626906597273206,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.3943,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10209717601537704,
      "step": 455,
      "valid_targets_mean": 4135.4,
      "valid_targets_min": 781
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.27504301015565424,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.4126,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09681200981140137,
      "step": 460,
      "valid_targets_mean": 4369.6,
      "valid_targets_min": 1097
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.25617518656405847,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.3957,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12154033780097961,
      "step": 465,
      "valid_targets_mean": 4883.8,
      "valid_targets_min": 1898
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.26931996681458753,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.391,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09775935113430023,
      "step": 470,
      "valid_targets_mean": 3897.2,
      "valid_targets_min": 622
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.26666715171044425,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.3771,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12007102370262146,
      "step": 475,
      "valid_targets_mean": 4395.6,
      "valid_targets_min": 919
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.2658064422635951,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.3875,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08204793930053711,
      "step": 480,
      "valid_targets_mean": 3838.0,
      "valid_targets_min": 677
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.25545083038676303,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.3895,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08605055510997772,
      "step": 485,
      "valid_targets_mean": 3645.2,
      "valid_targets_min": 555
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.25973106703348975,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.3885,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09146520495414734,
      "step": 490,
      "valid_targets_mean": 3420.6,
      "valid_targets_min": 535
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.2750742216517568,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.3764,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08960846066474915,
      "step": 495,
      "valid_targets_mean": 4136.1,
      "valid_targets_min": 655
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.2763712825473007,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.3843,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07676543295383453,
      "step": 500,
      "valid_targets_mean": 3452.6,
      "valid_targets_min": 843
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.27758734369621985,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3923,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08084428310394287,
      "step": 505,
      "valid_targets_mean": 3238.7,
      "valid_targets_min": 769
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.27049684231659876,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3774,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10289500653743744,
      "step": 510,
      "valid_targets_mean": 5543.2,
      "valid_targets_min": 2362
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.2909743818914385,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.3927,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12213883548974991,
      "step": 515,
      "valid_targets_mean": 4245.5,
      "valid_targets_min": 887
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.2612987981680651,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3859,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09372418373823166,
      "step": 520,
      "valid_targets_mean": 4074.0,
      "valid_targets_min": 744
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.2540523872893174,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.3892,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09320654720067978,
      "step": 525,
      "valid_targets_mean": 4009.7,
      "valid_targets_min": 692
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.28902543089403615,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3976,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11859594285488129,
      "step": 530,
      "valid_targets_mean": 4288.0,
      "valid_targets_min": 654
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.24825254456467905,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.3826,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07744468003511429,
      "step": 535,
      "valid_targets_mean": 3677.9,
      "valid_targets_min": 660
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.25942451584851345,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.3868,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08802760392427444,
      "step": 540,
      "valid_targets_mean": 3556.9,
      "valid_targets_min": 616
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.26827502593766855,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.3955,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10362309217453003,
      "step": 545,
      "valid_targets_mean": 3917.4,
      "valid_targets_min": 800
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.27104308274273714,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.3853,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10550689697265625,
      "step": 550,
      "valid_targets_mean": 4820.1,
      "valid_targets_min": 712
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.26893792764640523,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3855,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10886234790086746,
      "step": 555,
      "valid_targets_mean": 4253.8,
      "valid_targets_min": 885
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.25930526404440535,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.3775,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0945378988981247,
      "step": 560,
      "valid_targets_mean": 3958.1,
      "valid_targets_min": 659
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.25638387737684076,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.39,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08409561961889267,
      "step": 565,
      "valid_targets_mean": 3962.6,
      "valid_targets_min": 610
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.2671487792321955,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3825,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0818827822804451,
      "step": 570,
      "valid_targets_mean": 3622.9,
      "valid_targets_min": 906
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.26000255203013317,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.3856,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08918525278568268,
      "step": 575,
      "valid_targets_mean": 3669.9,
      "valid_targets_min": 525
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.2574886881135314,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.3771,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10702772438526154,
      "step": 580,
      "valid_targets_mean": 4520.2,
      "valid_targets_min": 741
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.25144794853906877,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.3814,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07891513407230377,
      "step": 585,
      "valid_targets_mean": 3333.7,
      "valid_targets_min": 485
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.2705475034417514,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3889,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0964181050658226,
      "step": 590,
      "valid_targets_mean": 3917.1,
      "valid_targets_min": 537
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.27596223847527274,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.3788,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1040067970752716,
      "step": 595,
      "valid_targets_mean": 4070.1,
      "valid_targets_min": 1130
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.24774417542704746,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.3852,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09477730095386505,
      "step": 600,
      "valid_targets_mean": 3849.6,
      "valid_targets_min": 825
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.28587902399434467,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3969,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0990772545337677,
      "step": 605,
      "valid_targets_mean": 4365.7,
      "valid_targets_min": 827
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.2435978738436119,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.3977,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09892106056213379,
      "step": 610,
      "valid_targets_mean": 4498.3,
      "valid_targets_min": 837
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.27062600457030334,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.3891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09334753453731537,
      "step": 615,
      "valid_targets_mean": 3371.8,
      "valid_targets_min": 684
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.2599279504730553,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.3981,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08131345361471176,
      "step": 620,
      "valid_targets_mean": 3058.2,
      "valid_targets_min": 721
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.2305136982364218,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.3915,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08032631874084473,
      "step": 625,
      "valid_targets_mean": 3987.6,
      "valid_targets_min": 822
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.2368983513491616,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3838,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1091882586479187,
      "step": 630,
      "valid_targets_mean": 4481.3,
      "valid_targets_min": 777
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.24655583186551253,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.3749,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09838627278804779,
      "step": 635,
      "valid_targets_mean": 5421.4,
      "valid_targets_min": 1439
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.2581095958964102,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.3815,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09013497829437256,
      "step": 640,
      "valid_targets_mean": 3464.9,
      "valid_targets_min": 684
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.2640398665017282,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.3856,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10049368441104889,
      "step": 645,
      "valid_targets_mean": 4081.4,
      "valid_targets_min": 740
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.27125424047590235,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.38,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08673174679279327,
      "step": 650,
      "valid_targets_mean": 3122.5,
      "valid_targets_min": 485
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.26866654993549727,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3922,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06575755774974823,
      "step": 655,
      "valid_targets_mean": 2845.2,
      "valid_targets_min": 686
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.25607904518963737,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.3791,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10241270065307617,
      "step": 660,
      "valid_targets_mean": 4285.4,
      "valid_targets_min": 834
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.25216108900772355,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.375,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08354803919792175,
      "step": 665,
      "valid_targets_mean": 3571.8,
      "valid_targets_min": 645
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.24227154891418914,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3781,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07896874099969864,
      "step": 670,
      "valid_targets_mean": 3512.8,
      "valid_targets_min": 776
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.24681324443358604,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3798,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08597329258918762,
      "step": 675,
      "valid_targets_mean": 3569.9,
      "valid_targets_min": 485
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.26441608498991254,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3784,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12063826620578766,
      "step": 680,
      "valid_targets_mean": 4745.7,
      "valid_targets_min": 1243
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.25528886372031545,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.385,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08705770969390869,
      "step": 685,
      "valid_targets_mean": 3251.2,
      "valid_targets_min": 1094
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.29374463130070577,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3929,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09442579001188278,
      "step": 690,
      "valid_targets_mean": 3977.1,
      "valid_targets_min": 581
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.2604822052363546,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.3811,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11611970514059067,
      "step": 695,
      "valid_targets_mean": 4396.9,
      "valid_targets_min": 1059
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.22409515043480815,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.3835,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1003444492816925,
      "step": 700,
      "valid_targets_mean": 4903.1,
      "valid_targets_min": 1373
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.2608751853365237,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3805,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10820289701223373,
      "step": 705,
      "valid_targets_mean": 3952.2,
      "valid_targets_min": 721
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.2528198345647317,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.3798,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09916079789400101,
      "step": 710,
      "valid_targets_mean": 3761.2,
      "valid_targets_min": 574
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.25819787404699024,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.3863,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09638336300849915,
      "step": 715,
      "valid_targets_mean": 4596.9,
      "valid_targets_min": 1016
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.2660494679823855,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3894,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08017449080944061,
      "step": 720,
      "valid_targets_mean": 3175.8,
      "valid_targets_min": 837
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.24110931712685146,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.3743,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09577009081840515,
      "step": 725,
      "valid_targets_mean": 4714.4,
      "valid_targets_min": 743
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.2598267703384784,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.3834,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09547718614339828,
      "step": 730,
      "valid_targets_mean": 4015.2,
      "valid_targets_min": 642
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.2452273156396588,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.3894,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09247712790966034,
      "step": 735,
      "valid_targets_mean": 3852.4,
      "valid_targets_min": 810
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.2727806967117683,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3812,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07677747309207916,
      "step": 740,
      "valid_targets_mean": 3152.3,
      "valid_targets_min": 576
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.2426475499348263,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.3738,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0934973657131195,
      "step": 745,
      "valid_targets_mean": 3767.7,
      "valid_targets_min": 1144
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.25729643915994693,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.3831,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08878760039806366,
      "step": 750,
      "valid_targets_mean": 3209.7,
      "valid_targets_min": 702
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.2791733106897808,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.3902,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11280880868434906,
      "step": 755,
      "valid_targets_mean": 4304.2,
      "valid_targets_min": 752
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.24742420092409653,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3768,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07657507061958313,
      "step": 760,
      "valid_targets_mean": 3594.1,
      "valid_targets_min": 660
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.24507167278237849,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3716,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1008102297782898,
      "step": 765,
      "valid_targets_mean": 4508.9,
      "valid_targets_min": 1312
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.2735281806020755,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.379,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08904020488262177,
      "step": 770,
      "valid_targets_mean": 2916.2,
      "valid_targets_min": 850
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.28818502912320515,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.3903,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09894673526287079,
      "step": 775,
      "valid_targets_mean": 4380.6,
      "valid_targets_min": 508
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.2545017444496391,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3878,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11805623769760132,
      "step": 780,
      "valid_targets_mean": 4306.9,
      "valid_targets_min": 816
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5313212494908685,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3896,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.39344656467437744,
      "step": 785,
      "valid_targets_mean": 3915.7,
      "valid_targets_min": 904
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.39344656467437744,
      "step": 785,
      "total_flos": 1.3428870206613094e+18,
      "train_loss": 0.4188855086162591,
      "train_runtime": 17401.9251,
      "train_samples_per_second": 2.873,
      "train_steps_per_second": 0.045,
      "valid_targets_mean": 3915.7,
      "valid_targets_min": 904
    }
  ],
  "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": 1.3428870206613094e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}