{
  "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": 5.205500278379217,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.6983,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.19261837005615234,
      "step": 5,
      "valid_targets_mean": 5115.1,
      "valid_targets_min": 721
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.5748338506783104,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.6766,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12646880745887756,
      "step": 10,
      "valid_targets_mean": 3549.1,
      "valid_targets_min": 658
    },
    {
      "epoch": 0.096,
      "grad_norm": 1.9948939845634819,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.635,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.20435824990272522,
      "step": 15,
      "valid_targets_mean": 6176.2,
      "valid_targets_min": 1474
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.8281715432344015,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.6013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15443837642669678,
      "step": 20,
      "valid_targets_mean": 5752.4,
      "valid_targets_min": 893
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.735094107252598,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.5839,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15552130341529846,
      "step": 25,
      "valid_targets_mean": 5077.4,
      "valid_targets_min": 696
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.555856049334831,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.5507,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12468452006578445,
      "step": 30,
      "valid_targets_mean": 3350.6,
      "valid_targets_min": 682
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.4367800673203073,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5135,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1386324167251587,
      "step": 35,
      "valid_targets_mean": 3797.4,
      "valid_targets_min": 993
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.4144521032218668,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.4951,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13024786114692688,
      "step": 40,
      "valid_targets_mean": 3820.9,
      "valid_targets_min": 632
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.28950453577240487,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.496,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.143850177526474,
      "step": 45,
      "valid_targets_mean": 7638.8,
      "valid_targets_min": 879
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3162141844456686,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.4735,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10481979697942734,
      "step": 50,
      "valid_targets_mean": 5154.4,
      "valid_targets_min": 969
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.3008478747607178,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.4645,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.129011869430542,
      "step": 55,
      "valid_targets_mean": 4607.6,
      "valid_targets_min": 611
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.33473945541498157,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.4479,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11655284464359283,
      "step": 60,
      "valid_targets_mean": 3424.2,
      "valid_targets_min": 699
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.2805147264282592,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.44,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09770642220973969,
      "step": 65,
      "valid_targets_mean": 4424.4,
      "valid_targets_min": 537
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.2816332250539912,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.4345,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13446155190467834,
      "step": 70,
      "valid_targets_mean": 4172.4,
      "valid_targets_min": 914
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2684061302848235,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.4324,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12554392218589783,
      "step": 75,
      "valid_targets_mean": 4814.1,
      "valid_targets_min": 1079
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.2547415620839725,
      "learning_rate": 4e-05,
      "loss": 0.4331,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10385149717330933,
      "step": 80,
      "valid_targets_mean": 3606.3,
      "valid_targets_min": 695
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.3127882084115677,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.4202,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11602122336626053,
      "step": 85,
      "valid_targets_mean": 3818.8,
      "valid_targets_min": 684
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.29263524972027855,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.4393,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1139579564332962,
      "step": 90,
      "valid_targets_mean": 3442.2,
      "valid_targets_min": 1222
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.2624148934409706,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4241,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12687838077545166,
      "step": 95,
      "valid_targets_mean": 4698.9,
      "valid_targets_min": 626
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3069586756830426,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.4372,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11603105068206787,
      "step": 100,
      "valid_targets_mean": 5901.6,
      "valid_targets_min": 1011
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.2887370334147458,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.4287,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10971199721097946,
      "step": 105,
      "valid_targets_mean": 3635.2,
      "valid_targets_min": 661
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.28521956009741617,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4267,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11025803536176682,
      "step": 110,
      "valid_targets_mean": 4132.8,
      "valid_targets_min": 727
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.27919261979244836,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.4294,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11331288516521454,
      "step": 115,
      "valid_targets_mean": 4321.4,
      "valid_targets_min": 1226
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.28478517485059646,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.4092,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10387001931667328,
      "step": 120,
      "valid_targets_mean": 3813.3,
      "valid_targets_min": 762
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.26609570517664816,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4156,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10383210331201553,
      "step": 125,
      "valid_targets_mean": 3997.2,
      "valid_targets_min": 690
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.2556008867336422,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.4046,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09731249511241913,
      "step": 130,
      "valid_targets_mean": 4166.9,
      "valid_targets_min": 672
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.3109905570449353,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.407,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14275780320167542,
      "step": 135,
      "valid_targets_mean": 4863.6,
      "valid_targets_min": 1005
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.2774630032183097,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.405,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11254678666591644,
      "step": 140,
      "valid_targets_mean": 4343.1,
      "valid_targets_min": 678
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.2734041697626819,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.3918,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08156347274780273,
      "step": 145,
      "valid_targets_mean": 4243.1,
      "valid_targets_min": 671
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.29887742915541465,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.4084,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08462947607040405,
      "step": 150,
      "valid_targets_mean": 3060.8,
      "valid_targets_min": 829
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.2790585150718942,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.4121,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13514775037765503,
      "step": 155,
      "valid_targets_mean": 5204.2,
      "valid_targets_min": 543
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.2524024593121093,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.4209,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11746951937675476,
      "step": 160,
      "valid_targets_mean": 4534.9,
      "valid_targets_min": 1332
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.2800884303720555,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.4038,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10607901960611343,
      "step": 165,
      "valid_targets_mean": 4443.5,
      "valid_targets_min": 942
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.27803620188741157,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.3821,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07709759473800659,
      "step": 170,
      "valid_targets_mean": 2840.8,
      "valid_targets_min": 784
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.32040752453923044,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.3961,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1191687136888504,
      "step": 175,
      "valid_targets_mean": 3540.4,
      "valid_targets_min": 676
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.2956442527622805,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.3897,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12804493308067322,
      "step": 180,
      "valid_targets_mean": 4185.3,
      "valid_targets_min": 865
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.26995625394805933,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.3979,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10453646630048752,
      "step": 185,
      "valid_targets_mean": 4691.3,
      "valid_targets_min": 672
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.27593660120860486,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.3909,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11954151093959808,
      "step": 190,
      "valid_targets_mean": 4940.8,
      "valid_targets_min": 762
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.27293652610142033,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.3727,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09296149760484695,
      "step": 195,
      "valid_targets_mean": 3906.0,
      "valid_targets_min": 679
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.29342476651167615,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.3873,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12881578505039215,
      "step": 200,
      "valid_targets_mean": 4605.4,
      "valid_targets_min": 796
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.31280748525377855,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.3771,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1101343035697937,
      "step": 205,
      "valid_targets_mean": 4544.1,
      "valid_targets_min": 1022
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.32614176044891724,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.3823,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07670246809720993,
      "step": 210,
      "valid_targets_mean": 2746.2,
      "valid_targets_min": 667
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.27245974978382576,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.3879,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09793829917907715,
      "step": 215,
      "valid_targets_mean": 4857.7,
      "valid_targets_min": 729
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.2866686854498495,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.3874,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10090164840221405,
      "step": 220,
      "valid_targets_mean": 4517.0,
      "valid_targets_min": 933
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.2772724463406222,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.3946,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11227937042713165,
      "step": 225,
      "valid_targets_mean": 4532.2,
      "valid_targets_min": 722
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.29296884033651444,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.4002,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09516061842441559,
      "step": 230,
      "valid_targets_mean": 3938.2,
      "valid_targets_min": 697
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.32421955078067777,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.3962,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0727391242980957,
      "step": 235,
      "valid_targets_mean": 2704.8,
      "valid_targets_min": 762
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.2816723753872089,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.3854,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09001927077770233,
      "step": 240,
      "valid_targets_mean": 3603.8,
      "valid_targets_min": 426
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.2964520827004003,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.3901,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06649242341518402,
      "step": 245,
      "valid_targets_mean": 2444.6,
      "valid_targets_min": 490
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.3246562610571258,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.3838,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07096904516220093,
      "step": 250,
      "valid_targets_mean": 3515.6,
      "valid_targets_min": 796
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.35858995829869394,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.3924,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10180254280567169,
      "step": 255,
      "valid_targets_mean": 2299.6,
      "valid_targets_min": 660
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.28079793467321496,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.3766,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11257487535476685,
      "step": 260,
      "valid_targets_mean": 3868.0,
      "valid_targets_min": 563
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.261936162427745,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.3949,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12114608287811279,
      "step": 265,
      "valid_targets_mean": 6315.8,
      "valid_targets_min": 702
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.2751370170577663,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.3765,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07566531002521515,
      "step": 270,
      "valid_targets_mean": 3108.9,
      "valid_targets_min": 922
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.27730223682709704,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.3804,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09624980390071869,
      "step": 275,
      "valid_targets_mean": 3650.1,
      "valid_targets_min": 656
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.2926894727747471,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.389,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09903205931186676,
      "step": 280,
      "valid_targets_mean": 4775.1,
      "valid_targets_min": 578
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.27679352038026606,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.3979,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10703089833259583,
      "step": 285,
      "valid_targets_mean": 4400.2,
      "valid_targets_min": 843
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.26672425055483295,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.3736,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09300033748149872,
      "step": 290,
      "valid_targets_mean": 4499.8,
      "valid_targets_min": 820
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.29555102671875205,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.375,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09723745286464691,
      "step": 295,
      "valid_targets_mean": 4538.1,
      "valid_targets_min": 695
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.2668281855131916,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.3785,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10853414982557297,
      "step": 300,
      "valid_targets_mean": 4488.4,
      "valid_targets_min": 1016
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.3101905842132621,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.3833,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08487877249717712,
      "step": 305,
      "valid_targets_mean": 2903.1,
      "valid_targets_min": 687
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.30540632355320063,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.375,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1010449007153511,
      "step": 310,
      "valid_targets_mean": 3098.8,
      "valid_targets_min": 1084
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.2860028168001083,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.3829,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.052919719368219376,
      "step": 315,
      "valid_targets_mean": 1940.7,
      "valid_targets_min": 506
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.26777128899987784,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.3686,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.052187301218509674,
      "step": 320,
      "valid_targets_mean": 2988.4,
      "valid_targets_min": 697
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.31112856211166856,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.3619,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11293686181306839,
      "step": 325,
      "valid_targets_mean": 4691.6,
      "valid_targets_min": 1015
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.2799869787275962,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.3629,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07240612804889679,
      "step": 330,
      "valid_targets_mean": 3009.1,
      "valid_targets_min": 671
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.2884308503608875,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.357,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08298136293888092,
      "step": 335,
      "valid_targets_mean": 4054.4,
      "valid_targets_min": 568
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.2844717924785759,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.3763,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10223925858736038,
      "step": 340,
      "valid_targets_mean": 4150.5,
      "valid_targets_min": 982
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.26298527961696344,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.3759,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11003424972295761,
      "step": 345,
      "valid_targets_mean": 4974.8,
      "valid_targets_min": 1081
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.2410374504916366,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.362,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10497504472732544,
      "step": 350,
      "valid_targets_mean": 6649.9,
      "valid_targets_min": 969
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.2724185465521075,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.3586,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08595755696296692,
      "step": 355,
      "valid_targets_mean": 4010.1,
      "valid_targets_min": 734
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.3126202375664361,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.3856,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12328476458787918,
      "step": 360,
      "valid_targets_mean": 3361.4,
      "valid_targets_min": 582
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.3314576218239427,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.3772,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10273909568786621,
      "step": 365,
      "valid_targets_mean": 4295.8,
      "valid_targets_min": 667
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.288146966623846,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.3749,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11177492886781693,
      "step": 370,
      "valid_targets_mean": 5310.6,
      "valid_targets_min": 662
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.27461558671949216,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3721,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08605611324310303,
      "step": 375,
      "valid_targets_mean": 4217.3,
      "valid_targets_min": 800
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.29561446642122985,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.3808,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11280474066734314,
      "step": 380,
      "valid_targets_mean": 4742.1,
      "valid_targets_min": 959
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.30351003633129336,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.3744,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07686591148376465,
      "step": 385,
      "valid_targets_mean": 2600.0,
      "valid_targets_min": 706
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.29272737837051693,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.3643,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11120889335870743,
      "step": 390,
      "valid_targets_mean": 4677.1,
      "valid_targets_min": 658
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.2554948982620788,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.3701,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09617634862661362,
      "step": 395,
      "valid_targets_mean": 4526.2,
      "valid_targets_min": 810
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.28652356910442905,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.3701,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10789115726947784,
      "step": 400,
      "valid_targets_mean": 4557.2,
      "valid_targets_min": 668
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.28834266263374586,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.3807,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08078165352344513,
      "step": 405,
      "valid_targets_mean": 3316.8,
      "valid_targets_min": 693
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.2865343141685818,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.3592,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06548111140727997,
      "step": 410,
      "valid_targets_mean": 3193.4,
      "valid_targets_min": 549
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.29513828345428855,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.3831,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10732559114694595,
      "step": 415,
      "valid_targets_mean": 2967.9,
      "valid_targets_min": 546
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.2750256063229265,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.3694,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10158859193325043,
      "step": 420,
      "valid_targets_mean": 4297.9,
      "valid_targets_min": 735
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.25055723900490434,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.3568,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09696809947490692,
      "step": 425,
      "valid_targets_mean": 5163.2,
      "valid_targets_min": 780
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.279998257547057,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.3574,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09604936093091965,
      "step": 430,
      "valid_targets_mean": 3806.9,
      "valid_targets_min": 823
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.2655986995368152,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.3617,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10295376181602478,
      "step": 435,
      "valid_targets_mean": 4629.6,
      "valid_targets_min": 575
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.27360232283299474,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3585,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0704706534743309,
      "step": 440,
      "valid_targets_mean": 2763.1,
      "valid_targets_min": 599
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.2751327965704248,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.3637,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07666494697332382,
      "step": 445,
      "valid_targets_mean": 3470.4,
      "valid_targets_min": 523
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.3080456403977746,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3737,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07470028102397919,
      "step": 450,
      "valid_targets_mean": 2147.8,
      "valid_targets_min": 888
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.2624318191343192,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.3712,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11898545920848846,
      "step": 455,
      "valid_targets_mean": 5307.3,
      "valid_targets_min": 676
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.24619442355842874,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.3551,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.091619111597538,
      "step": 460,
      "valid_targets_mean": 4588.8,
      "valid_targets_min": 656
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.264791886786886,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.3697,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12300880253314972,
      "step": 465,
      "valid_targets_mean": 5310.1,
      "valid_targets_min": 1026
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.2370783908772975,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.3604,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09207291156053543,
      "step": 470,
      "valid_targets_mean": 6178.0,
      "valid_targets_min": 766
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.30391113301450906,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.3589,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10276287794113159,
      "step": 475,
      "valid_targets_mean": 4018.2,
      "valid_targets_min": 701
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.28350565619327855,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.354,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09772688895463943,
      "step": 480,
      "valid_targets_mean": 3783.2,
      "valid_targets_min": 613
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.26052072655453223,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.3738,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09954036772251129,
      "step": 485,
      "valid_targets_mean": 5202.6,
      "valid_targets_min": 846
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.23891779914584613,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.351,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07682588696479797,
      "step": 490,
      "valid_targets_mean": 5238.2,
      "valid_targets_min": 615
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.28975386864608443,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.3691,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10776814818382263,
      "step": 495,
      "valid_targets_mean": 4433.8,
      "valid_targets_min": 705
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.29232169027260746,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.3475,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11693393439054489,
      "step": 500,
      "valid_targets_mean": 4741.5,
      "valid_targets_min": 956
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.26670877940973337,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3506,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0831618681550026,
      "step": 505,
      "valid_targets_mean": 3530.6,
      "valid_targets_min": 905
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.31969473635944046,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3878,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12326958775520325,
      "step": 510,
      "valid_targets_mean": 3617.4,
      "valid_targets_min": 754
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.2508568809127141,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.3552,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08297090232372284,
      "step": 515,
      "valid_targets_mean": 5496.9,
      "valid_targets_min": 556
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.2655695833704515,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3524,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05646228790283203,
      "step": 520,
      "valid_targets_mean": 2416.6,
      "valid_targets_min": 701
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.23635852470839794,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.3496,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08186306059360504,
      "step": 525,
      "valid_targets_mean": 4963.5,
      "valid_targets_min": 874
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.2561532435289236,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3596,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1043807715177536,
      "step": 530,
      "valid_targets_mean": 4564.0,
      "valid_targets_min": 572
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.3004178203722905,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.3688,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09910893440246582,
      "step": 535,
      "valid_targets_mean": 3899.0,
      "valid_targets_min": 656
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.25545226401555293,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.3637,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11958801001310349,
      "step": 540,
      "valid_targets_mean": 5934.1,
      "valid_targets_min": 1265
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.25619643151753957,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.3459,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06787975132465363,
      "step": 545,
      "valid_targets_mean": 4528.9,
      "valid_targets_min": 667
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.2728042791546083,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.3382,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.04455628991127014,
      "step": 550,
      "valid_targets_mean": 1732.6,
      "valid_targets_min": 600
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.26363753539691687,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3775,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10106958448886871,
      "step": 555,
      "valid_targets_mean": 4303.4,
      "valid_targets_min": 833
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.2550059778936495,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.353,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07427558302879333,
      "step": 560,
      "valid_targets_mean": 4853.2,
      "valid_targets_min": 774
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.23808935752505023,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.3534,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09370896965265274,
      "step": 565,
      "valid_targets_mean": 6003.7,
      "valid_targets_min": 897
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.29099776938311844,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3507,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07799384742975235,
      "step": 570,
      "valid_targets_mean": 3052.6,
      "valid_targets_min": 1002
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.28573593311370926,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.3622,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09398327767848969,
      "step": 575,
      "valid_targets_mean": 3977.4,
      "valid_targets_min": 543
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.2692704279070834,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.349,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08807209134101868,
      "step": 580,
      "valid_targets_mean": 3948.5,
      "valid_targets_min": 804
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.24324903140753096,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.3737,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09824167937040329,
      "step": 585,
      "valid_targets_mean": 5340.4,
      "valid_targets_min": 1117
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.24430754008009728,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3418,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10011763125658035,
      "step": 590,
      "valid_targets_mean": 4848.2,
      "valid_targets_min": 894
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.2626496890552,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.3628,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0938146784901619,
      "step": 595,
      "valid_targets_mean": 4312.2,
      "valid_targets_min": 430
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.23972064848558458,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.366,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11136816442012787,
      "step": 600,
      "valid_targets_mean": 6651.5,
      "valid_targets_min": 1086
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.26704051965793485,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3431,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08306808024644852,
      "step": 605,
      "valid_targets_mean": 4450.2,
      "valid_targets_min": 745
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.25787116809210525,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.3533,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0934668555855751,
      "step": 610,
      "valid_targets_mean": 4839.4,
      "valid_targets_min": 832
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.25255278567050754,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.3518,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09989595413208008,
      "step": 615,
      "valid_targets_mean": 4687.4,
      "valid_targets_min": 585
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.2294145094914024,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.364,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09933659434318542,
      "step": 620,
      "valid_targets_mean": 4801.4,
      "valid_targets_min": 789
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.26465817723595647,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.3524,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0769161731004715,
      "step": 625,
      "valid_targets_mean": 3914.4,
      "valid_targets_min": 914
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.25688649927777807,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3602,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08550512045621872,
      "step": 630,
      "valid_targets_mean": 3864.8,
      "valid_targets_min": 685
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.2535517653120582,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.3553,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07645588368177414,
      "step": 635,
      "valid_targets_mean": 3080.2,
      "valid_targets_min": 717
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.26559203871118275,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.3379,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08110323548316956,
      "step": 640,
      "valid_targets_mean": 3450.9,
      "valid_targets_min": 678
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.2800456759137408,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.3474,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09628013521432877,
      "step": 645,
      "valid_targets_mean": 4791.2,
      "valid_targets_min": 676
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.3176910180326539,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.3688,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09897404164075851,
      "step": 650,
      "valid_targets_mean": 3084.0,
      "valid_targets_min": 660
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.3127436987611351,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3544,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08276864141225815,
      "step": 655,
      "valid_targets_mean": 3368.8,
      "valid_targets_min": 656
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.2615076074985247,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.3726,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08330519497394562,
      "step": 660,
      "valid_targets_mean": 4624.8,
      "valid_targets_min": 905
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.2645729908704741,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.3535,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11851125955581665,
      "step": 665,
      "valid_targets_mean": 4883.4,
      "valid_targets_min": 521
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.29012788912817494,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3531,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08992412686347961,
      "step": 670,
      "valid_targets_mean": 3362.1,
      "valid_targets_min": 588
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.3178767162911069,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3573,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07908840477466583,
      "step": 675,
      "valid_targets_mean": 2647.6,
      "valid_targets_min": 665
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.25273126562165116,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3375,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0724274218082428,
      "step": 680,
      "valid_targets_mean": 3349.6,
      "valid_targets_min": 1064
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.27465343832678074,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.3487,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11378444731235504,
      "step": 685,
      "valid_targets_mean": 5737.0,
      "valid_targets_min": 672
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.2561924911459456,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3366,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06337870657444,
      "step": 690,
      "valid_targets_mean": 3790.9,
      "valid_targets_min": 742
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.2641603105306008,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.352,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09154815971851349,
      "step": 695,
      "valid_targets_mean": 3475.6,
      "valid_targets_min": 1003
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.2428174273861076,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.3591,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08129976689815521,
      "step": 700,
      "valid_targets_mean": 3895.2,
      "valid_targets_min": 626
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.21278653648741958,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3517,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05930912494659424,
      "step": 705,
      "valid_targets_mean": 4292.9,
      "valid_targets_min": 705
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.27020471881967667,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.3455,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08879420161247253,
      "step": 710,
      "valid_targets_mean": 3785.3,
      "valid_targets_min": 909
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.2591359154854987,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.3586,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09846097230911255,
      "step": 715,
      "valid_targets_mean": 4014.4,
      "valid_targets_min": 714
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.2914672243018568,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3597,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0793735459446907,
      "step": 720,
      "valid_targets_mean": 3110.8,
      "valid_targets_min": 668
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.32708485868699294,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.3614,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10227176547050476,
      "step": 725,
      "valid_targets_mean": 3199.5,
      "valid_targets_min": 523
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.2627009111629682,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.3544,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05896441638469696,
      "step": 730,
      "valid_targets_mean": 3711.3,
      "valid_targets_min": 556
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.257043393006274,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.3685,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09447735548019409,
      "step": 735,
      "valid_targets_mean": 4783.8,
      "valid_targets_min": 942
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.3997471701885391,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3432,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09453587234020233,
      "step": 740,
      "valid_targets_mean": 3679.1,
      "valid_targets_min": 743
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.26864672730907413,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.3487,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07397910952568054,
      "step": 745,
      "valid_targets_mean": 3144.4,
      "valid_targets_min": 571
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.2647787720702242,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.343,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09614728391170502,
      "step": 750,
      "valid_targets_mean": 4536.8,
      "valid_targets_min": 699
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.2658402740359253,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.363,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08994534611701965,
      "step": 755,
      "valid_targets_mean": 3964.2,
      "valid_targets_min": 948
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.26992265781297736,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3639,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10105564445257187,
      "step": 760,
      "valid_targets_mean": 4507.9,
      "valid_targets_min": 816
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.2629295171576091,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3427,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07084855437278748,
      "step": 765,
      "valid_targets_mean": 3377.8,
      "valid_targets_min": 527
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.27633882847942576,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.3405,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0791630893945694,
      "step": 770,
      "valid_targets_mean": 3862.8,
      "valid_targets_min": 746
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.27980295029529056,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.3681,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14825886487960815,
      "step": 775,
      "valid_targets_mean": 5824.7,
      "valid_targets_min": 673
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.23185972022172793,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3322,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07116097956895828,
      "step": 780,
      "valid_targets_mean": 4343.6,
      "valid_targets_min": 909
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.4303965779825097,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3628,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.3454465866088867,
      "step": 785,
      "valid_targets_mean": 6003.4,
      "valid_targets_min": 835
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.3454465866088867,
      "step": 785,
      "total_flos": 1.0904976209980948e+18,
      "train_loss": 0.3874767948867409,
      "train_runtime": 16769.139,
      "train_samples_per_second": 2.982,
      "train_steps_per_second": 0.047,
      "valid_targets_mean": 6003.4,
      "valid_targets_min": 835
    }
  ],
  "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": 1.0904976209980948e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}