{
  "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": 4.925816001371351,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.6621,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15316620469093323,
      "step": 5,
      "valid_targets_mean": 5615.4,
      "valid_targets_min": 1301
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.6038645425133113,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.6885,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1408233940601349,
      "step": 10,
      "valid_targets_mean": 5344.8,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.096,
      "grad_norm": 1.9219967246450544,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.6124,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16160334646701813,
      "step": 15,
      "valid_targets_mean": 5692.6,
      "valid_targets_min": 2447
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.9066524081071425,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.5859,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1370714008808136,
      "step": 20,
      "valid_targets_mean": 4999.8,
      "valid_targets_min": 834
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6858551047823068,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.5577,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1048346608877182,
      "step": 25,
      "valid_targets_mean": 3834.4,
      "valid_targets_min": 627
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.4737056753916451,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.5357,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11810168623924255,
      "step": 30,
      "valid_targets_mean": 5133.2,
      "valid_targets_min": 695
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.3987307487395899,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5225,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15136444568634033,
      "step": 35,
      "valid_targets_mean": 5572.9,
      "valid_targets_min": 1349
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.30408841177006257,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.4856,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09305448830127716,
      "step": 40,
      "valid_targets_mean": 5469.4,
      "valid_targets_min": 1156
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.28023584477307756,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.4699,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08279567956924438,
      "step": 45,
      "valid_targets_mean": 3771.6,
      "valid_targets_min": 505
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.2623299739038126,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.4638,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0865197628736496,
      "step": 50,
      "valid_targets_mean": 4588.6,
      "valid_targets_min": 891
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.23584052712257855,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.4584,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11883324384689331,
      "step": 55,
      "valid_targets_mean": 6565.4,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.22945467495730265,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.4345,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10670436173677444,
      "step": 60,
      "valid_targets_mean": 5431.0,
      "valid_targets_min": 748
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.25958034421769316,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.4568,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10006660968065262,
      "step": 65,
      "valid_targets_mean": 4435.2,
      "valid_targets_min": 666
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.25101114300254473,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.4288,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0953449234366417,
      "step": 70,
      "valid_targets_mean": 5001.8,
      "valid_targets_min": 770
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.21760950700440734,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.4386,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09250706434249878,
      "step": 75,
      "valid_targets_mean": 5227.5,
      "valid_targets_min": 442
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.26479413690674075,
      "learning_rate": 4e-05,
      "loss": 0.4307,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10058815777301788,
      "step": 80,
      "valid_targets_mean": 4494.2,
      "valid_targets_min": 712
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.2336713450215106,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.4131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0961797833442688,
      "step": 85,
      "valid_targets_mean": 6049.8,
      "valid_targets_min": 659
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.24198160764095675,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.4148,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10415308177471161,
      "step": 90,
      "valid_targets_mean": 4485.3,
      "valid_targets_min": 670
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.22408845086959975,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4156,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1032426655292511,
      "step": 95,
      "valid_targets_mean": 5694.4,
      "valid_targets_min": 974
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.36120176963915784,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.4079,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10340757668018341,
      "step": 100,
      "valid_targets_mean": 5096.6,
      "valid_targets_min": 886
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.24256095374026376,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.409,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12146667391061783,
      "step": 105,
      "valid_targets_mean": 6735.6,
      "valid_targets_min": 760
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.22862627049602116,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4045,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1170661598443985,
      "step": 110,
      "valid_targets_mean": 6922.1,
      "valid_targets_min": 1467
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.25198968443722136,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.4027,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11473138630390167,
      "step": 115,
      "valid_targets_mean": 5690.5,
      "valid_targets_min": 576
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.22248181013285756,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.4025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10501393675804138,
      "step": 120,
      "valid_targets_mean": 5778.5,
      "valid_targets_min": 1069
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2307629663338528,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4056,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08154725283384323,
      "step": 125,
      "valid_targets_mean": 5937.2,
      "valid_targets_min": 1019
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.23384759922389925,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.4039,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10147960484027863,
      "step": 130,
      "valid_targets_mean": 4729.8,
      "valid_targets_min": 687
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.23008776759865265,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.4101,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09820076823234558,
      "step": 135,
      "valid_targets_mean": 4614.9,
      "valid_targets_min": 496
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.23922899554829585,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.3919,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08367419987916946,
      "step": 140,
      "valid_targets_mean": 5070.4,
      "valid_targets_min": 729
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.2273633281215133,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.3775,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09413500875234604,
      "step": 145,
      "valid_targets_mean": 5735.8,
      "valid_targets_min": 812
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.26293258335241,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.3917,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11875556409358978,
      "step": 150,
      "valid_targets_mean": 5658.6,
      "valid_targets_min": 987
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.24870909247092857,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.3837,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09130781888961792,
      "step": 155,
      "valid_targets_mean": 5792.5,
      "valid_targets_min": 676
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.23656516386477458,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.3957,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09010981023311615,
      "step": 160,
      "valid_targets_mean": 4240.8,
      "valid_targets_min": 575
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.24052725165140795,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.3859,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09425537288188934,
      "step": 165,
      "valid_targets_mean": 5526.6,
      "valid_targets_min": 683
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.22871310720091415,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.3889,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08064226806163788,
      "step": 170,
      "valid_targets_mean": 4734.3,
      "valid_targets_min": 858
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.23207100949526205,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.378,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09184914827346802,
      "step": 175,
      "valid_targets_mean": 4519.6,
      "valid_targets_min": 535
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.2268464194505428,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.3799,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1001918688416481,
      "step": 180,
      "valid_targets_mean": 5724.2,
      "valid_targets_min": 1237
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.2565054263960601,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.383,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0865120217204094,
      "step": 185,
      "valid_targets_mean": 4131.6,
      "valid_targets_min": 826
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.20879271728086604,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.3821,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0937279611825943,
      "step": 190,
      "valid_targets_mean": 5937.2,
      "valid_targets_min": 735
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.23126357655080354,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.3713,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10395435243844986,
      "step": 195,
      "valid_targets_mean": 6906.5,
      "valid_targets_min": 2016
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.23717319419246105,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.3828,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09835539758205414,
      "step": 200,
      "valid_targets_mean": 5437.2,
      "valid_targets_min": 824
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.35968020546013296,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.3815,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07530041038990021,
      "step": 205,
      "valid_targets_mean": 5013.6,
      "valid_targets_min": 712
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.24046800998054815,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.3878,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11328396946191788,
      "step": 210,
      "valid_targets_mean": 6360.2,
      "valid_targets_min": 1310
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.24526162606684027,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.3818,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07932275533676147,
      "step": 215,
      "valid_targets_mean": 4206.3,
      "valid_targets_min": 544
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.281741811074157,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.3799,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10239493101835251,
      "step": 220,
      "valid_targets_mean": 6040.5,
      "valid_targets_min": 615
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.23918163077762844,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.377,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10365035384893417,
      "step": 225,
      "valid_targets_mean": 5706.7,
      "valid_targets_min": 747
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.2307808416847612,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.3814,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09817519783973694,
      "step": 230,
      "valid_targets_mean": 6541.8,
      "valid_targets_min": 1951
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.28294813209329395,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.3766,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.104357048869133,
      "step": 235,
      "valid_targets_mean": 5665.6,
      "valid_targets_min": 946
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.22957127344835665,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.3798,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07624883949756622,
      "step": 240,
      "valid_targets_mean": 4442.8,
      "valid_targets_min": 627
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.23157621260116973,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.3768,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09372389316558838,
      "step": 245,
      "valid_targets_mean": 6213.5,
      "valid_targets_min": 1380
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.251557645562299,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.3978,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09636591374874115,
      "step": 250,
      "valid_targets_mean": 6230.4,
      "valid_targets_min": 988
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.22282170912257812,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.3775,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07495754957199097,
      "step": 255,
      "valid_targets_mean": 5801.8,
      "valid_targets_min": 699
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.21437232283170424,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.3728,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09263433516025543,
      "step": 260,
      "valid_targets_mean": 6159.1,
      "valid_targets_min": 840
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.23003116022099923,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.374,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08770999312400818,
      "step": 265,
      "valid_targets_mean": 4599.5,
      "valid_targets_min": 717
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.22184131464205806,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.3655,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10612646490335464,
      "step": 270,
      "valid_targets_mean": 6200.3,
      "valid_targets_min": 834
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.232368163371399,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.3652,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09963122755289078,
      "step": 275,
      "valid_targets_mean": 5060.5,
      "valid_targets_min": 1147
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.23732321351995295,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.3745,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1050419732928276,
      "step": 280,
      "valid_targets_mean": 6499.1,
      "valid_targets_min": 1588
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.23570370078440928,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.3684,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09922470897436142,
      "step": 285,
      "valid_targets_mean": 5659.1,
      "valid_targets_min": 994
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.2181429073249209,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.372,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0864710733294487,
      "step": 290,
      "valid_targets_mean": 6055.6,
      "valid_targets_min": 672
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.22598138320758313,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.3775,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09190896153450012,
      "step": 295,
      "valid_targets_mean": 5374.9,
      "valid_targets_min": 805
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.2194343909194051,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.3711,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10858743637800217,
      "step": 300,
      "valid_targets_mean": 6487.9,
      "valid_targets_min": 718
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.21797248434187552,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.3697,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08281909674406052,
      "step": 305,
      "valid_targets_mean": 5484.2,
      "valid_targets_min": 605
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.2504302974310377,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.381,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08115041255950928,
      "step": 310,
      "valid_targets_mean": 4204.3,
      "valid_targets_min": 663
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.2278583979366395,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.3751,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09375721216201782,
      "step": 315,
      "valid_targets_mean": 6225.1,
      "valid_targets_min": 1309
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.2302462717851301,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.3752,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10975902527570724,
      "step": 320,
      "valid_targets_mean": 6777.1,
      "valid_targets_min": 842
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.2500340565679184,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.3667,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10280564427375793,
      "step": 325,
      "valid_targets_mean": 6622.0,
      "valid_targets_min": 962
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.21337742816465596,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.3569,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07776390016078949,
      "step": 330,
      "valid_targets_mean": 5986.1,
      "valid_targets_min": 631
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.2508936239712304,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.361,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0718369409441948,
      "step": 335,
      "valid_targets_mean": 4342.3,
      "valid_targets_min": 831
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.23955711979482625,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.3623,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11865168809890747,
      "step": 340,
      "valid_targets_mean": 6094.5,
      "valid_targets_min": 798
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.22032798211593765,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.3607,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08003910630941391,
      "step": 345,
      "valid_targets_mean": 6063.9,
      "valid_targets_min": 1315
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.23157157616125337,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.3621,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10399216413497925,
      "step": 350,
      "valid_targets_mean": 6594.4,
      "valid_targets_min": 773
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.22988855886392837,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.3612,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07251277565956116,
      "step": 355,
      "valid_targets_mean": 4493.4,
      "valid_targets_min": 779
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.25369176399383053,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.3603,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08884517848491669,
      "step": 360,
      "valid_targets_mean": 5150.9,
      "valid_targets_min": 720
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.23021442547864848,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.3695,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08386367559432983,
      "step": 365,
      "valid_targets_mean": 5094.7,
      "valid_targets_min": 1164
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.21827176704737583,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.3641,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09923890233039856,
      "step": 370,
      "valid_targets_mean": 6310.4,
      "valid_targets_min": 843
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.2142635583748864,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3635,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09996455907821655,
      "step": 375,
      "valid_targets_mean": 6558.8,
      "valid_targets_min": 475
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.24122556224556901,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.3649,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07641372084617615,
      "step": 380,
      "valid_targets_mean": 4674.8,
      "valid_targets_min": 724
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.22988548566979566,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.3597,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11224524676799774,
      "step": 385,
      "valid_targets_mean": 6512.4,
      "valid_targets_min": 1490
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.2382366640328879,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.3606,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08308979868888855,
      "step": 390,
      "valid_targets_mean": 4992.9,
      "valid_targets_min": 603
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.231316806080095,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.3641,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09834016114473343,
      "step": 395,
      "valid_targets_mean": 6120.9,
      "valid_targets_min": 1146
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.2158330304056028,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.3621,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06864991039037704,
      "step": 400,
      "valid_targets_mean": 4506.3,
      "valid_targets_min": 484
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.2366088070452868,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.3692,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08551353216171265,
      "step": 405,
      "valid_targets_mean": 5304.5,
      "valid_targets_min": 589
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.2286577480849222,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.365,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0804942399263382,
      "step": 410,
      "valid_targets_mean": 4903.5,
      "valid_targets_min": 861
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.23178373514592107,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.3636,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08426219969987869,
      "step": 415,
      "valid_targets_mean": 4515.4,
      "valid_targets_min": 615
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.23415850034103794,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.367,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09100905060768127,
      "step": 420,
      "valid_targets_mean": 5502.7,
      "valid_targets_min": 644
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.25788346438164567,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.3656,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10013468563556671,
      "step": 425,
      "valid_targets_mean": 7158.4,
      "valid_targets_min": 920
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.2663168450026528,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.3639,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10933417081832886,
      "step": 430,
      "valid_targets_mean": 5575.8,
      "valid_targets_min": 585
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.22657666928978396,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.3511,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0875961184501648,
      "step": 435,
      "valid_targets_mean": 5409.7,
      "valid_targets_min": 619
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.20762446157137557,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3531,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09684012830257416,
      "step": 440,
      "valid_targets_mean": 7491.9,
      "valid_targets_min": 997
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.2094923424751093,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.3654,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06848344206809998,
      "step": 445,
      "valid_targets_mean": 4123.4,
      "valid_targets_min": 972
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.238835533295405,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3622,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07796338200569153,
      "step": 450,
      "valid_targets_mean": 4773.1,
      "valid_targets_min": 535
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.23198080879921099,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.3505,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09783093631267548,
      "step": 455,
      "valid_targets_mean": 4947.8,
      "valid_targets_min": 956
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.22481005467458817,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.3586,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0951041504740715,
      "step": 460,
      "valid_targets_mean": 5319.2,
      "valid_targets_min": 494
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.21909144022779042,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.355,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08194763958454132,
      "step": 465,
      "valid_targets_mean": 6034.9,
      "valid_targets_min": 783
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.24361544866934678,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.3651,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09759137034416199,
      "step": 470,
      "valid_targets_mean": 5008.9,
      "valid_targets_min": 924
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.2016601334929828,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.3607,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10218866169452667,
      "step": 475,
      "valid_targets_mean": 7441.6,
      "valid_targets_min": 2449
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.23840750792843468,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.362,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09875436127185822,
      "step": 480,
      "valid_targets_mean": 6337.4,
      "valid_targets_min": 597
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.24239915747456078,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.349,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0895923301577568,
      "step": 485,
      "valid_targets_mean": 6322.4,
      "valid_targets_min": 839
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.21242777049065,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.3628,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10297973453998566,
      "step": 490,
      "valid_targets_mean": 6493.8,
      "valid_targets_min": 560
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.22847407513101906,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.361,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08375167101621628,
      "step": 495,
      "valid_targets_mean": 5043.3,
      "valid_targets_min": 926
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.24586460965020462,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.3576,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08625875413417816,
      "step": 500,
      "valid_targets_mean": 5513.8,
      "valid_targets_min": 761
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.2104796732926289,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3576,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09489600360393524,
      "step": 505,
      "valid_targets_mean": 6466.4,
      "valid_targets_min": 610
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.23165585722917922,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3541,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08994518220424652,
      "step": 510,
      "valid_targets_mean": 5873.0,
      "valid_targets_min": 580
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.22725684747524982,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.355,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0889921486377716,
      "step": 515,
      "valid_targets_mean": 4970.9,
      "valid_targets_min": 724
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.20619667973730407,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3597,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10102991759777069,
      "step": 520,
      "valid_targets_mean": 7098.2,
      "valid_targets_min": 1160
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.21917910359499615,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.344,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08251196891069412,
      "step": 525,
      "valid_targets_mean": 4624.6,
      "valid_targets_min": 617
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.20650291768137613,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3469,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08979316800832748,
      "step": 530,
      "valid_targets_mean": 6418.4,
      "valid_targets_min": 607
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.216094547753801,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.3544,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0932360589504242,
      "step": 535,
      "valid_targets_mean": 5597.5,
      "valid_targets_min": 760
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.21281946165230944,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.3568,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08613704890012741,
      "step": 540,
      "valid_targets_mean": 7236.2,
      "valid_targets_min": 855
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.2246368736111037,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.3461,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08083291351795197,
      "step": 545,
      "valid_targets_mean": 4563.4,
      "valid_targets_min": 631
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.2331858163667793,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.3553,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09777645766735077,
      "step": 550,
      "valid_targets_mean": 5753.4,
      "valid_targets_min": 1062
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.21721262332976055,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3559,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08036057651042938,
      "step": 555,
      "valid_targets_mean": 4847.0,
      "valid_targets_min": 787
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.24113837427031543,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.3615,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10509349405765533,
      "step": 560,
      "valid_targets_mean": 6220.2,
      "valid_targets_min": 544
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.3203309140334649,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.3511,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08810634911060333,
      "step": 565,
      "valid_targets_mean": 5272.6,
      "valid_targets_min": 762
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.21080834516529504,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3534,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08992394804954529,
      "step": 570,
      "valid_targets_mean": 5580.8,
      "valid_targets_min": 627
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.23999804676893177,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.3533,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11131718754768372,
      "step": 575,
      "valid_targets_mean": 6170.8,
      "valid_targets_min": 1551
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.22485406774402675,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.3585,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07756373286247253,
      "step": 580,
      "valid_targets_mean": 5085.8,
      "valid_targets_min": 965
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.21291429043493842,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.353,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09584388136863708,
      "step": 585,
      "valid_targets_mean": 5766.2,
      "valid_targets_min": 748
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.20777703805518813,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3413,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08502918481826782,
      "step": 590,
      "valid_targets_mean": 6093.1,
      "valid_targets_min": 949
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.20736032883078803,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.355,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07885292172431946,
      "step": 595,
      "valid_targets_mean": 5630.1,
      "valid_targets_min": 1145
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.21498651499573626,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.3556,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06810902059078217,
      "step": 600,
      "valid_targets_mean": 4383.6,
      "valid_targets_min": 673
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.21297617917135236,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3508,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07745738327503204,
      "step": 605,
      "valid_targets_mean": 4404.9,
      "valid_targets_min": 672
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.20937815919913053,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.3656,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10144443064928055,
      "step": 610,
      "valid_targets_mean": 6201.3,
      "valid_targets_min": 633
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.20961932864285193,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.3497,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08958933502435684,
      "step": 615,
      "valid_targets_mean": 6676.8,
      "valid_targets_min": 841
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.2132900632103475,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.3515,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09926494210958481,
      "step": 620,
      "valid_targets_mean": 6056.0,
      "valid_targets_min": 1621
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.21969664998278665,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.346,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09499312937259674,
      "step": 625,
      "valid_targets_mean": 6069.6,
      "valid_targets_min": 996
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.2271730718707198,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3445,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08944408595561981,
      "step": 630,
      "valid_targets_mean": 5608.9,
      "valid_targets_min": 1098
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.210249845105253,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.3524,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09703782945871353,
      "step": 635,
      "valid_targets_mean": 5320.3,
      "valid_targets_min": 610
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.21425761952188718,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.3517,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09806178510189056,
      "step": 640,
      "valid_targets_mean": 6156.8,
      "valid_targets_min": 1075
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.19296452759142185,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.3462,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08750709891319275,
      "step": 645,
      "valid_targets_mean": 5939.2,
      "valid_targets_min": 679
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.20817469362950297,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.3436,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06587532162666321,
      "step": 650,
      "valid_targets_mean": 4388.8,
      "valid_targets_min": 524
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.28860134218752004,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3488,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09858383238315582,
      "step": 655,
      "valid_targets_mean": 6164.9,
      "valid_targets_min": 1832
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.2274447884512814,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.351,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07915853708982468,
      "step": 660,
      "valid_targets_mean": 4616.7,
      "valid_targets_min": 900
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.22441955360635754,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.3523,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08973248302936554,
      "step": 665,
      "valid_targets_mean": 5245.9,
      "valid_targets_min": 689
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.22165020623426437,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3467,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07190017402172089,
      "step": 670,
      "valid_targets_mean": 4224.2,
      "valid_targets_min": 633
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.22088071426569952,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3362,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1008359044790268,
      "step": 675,
      "valid_targets_mean": 6355.0,
      "valid_targets_min": 814
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.19919434220748514,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3522,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08908769488334656,
      "step": 680,
      "valid_targets_mean": 6297.8,
      "valid_targets_min": 667
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.22342299709303468,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.353,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07984830439090729,
      "step": 685,
      "valid_targets_mean": 4236.8,
      "valid_targets_min": 810
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.21685501342432004,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3539,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09451751410961151,
      "step": 690,
      "valid_targets_mean": 5870.2,
      "valid_targets_min": 562
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.2288912352241636,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.3533,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10398273169994354,
      "step": 695,
      "valid_targets_mean": 5586.2,
      "valid_targets_min": 697
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.23156579703452698,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.3557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07878932356834412,
      "step": 700,
      "valid_targets_mean": 4835.4,
      "valid_targets_min": 693
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.21164725825560687,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3466,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09743297100067139,
      "step": 705,
      "valid_targets_mean": 6111.4,
      "valid_targets_min": 989
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.21414053041752096,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.3533,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10428131371736526,
      "step": 710,
      "valid_targets_mean": 6066.4,
      "valid_targets_min": 832
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.22327848120991073,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.3554,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10424737632274628,
      "step": 715,
      "valid_targets_mean": 5980.2,
      "valid_targets_min": 960
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.19247156612874025,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3457,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07965318858623505,
      "step": 720,
      "valid_targets_mean": 5630.6,
      "valid_targets_min": 1182
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.2104354863953018,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.357,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10120499134063721,
      "step": 725,
      "valid_targets_mean": 5889.6,
      "valid_targets_min": 1024
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.21319413636231446,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.3446,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07628591358661652,
      "step": 730,
      "valid_targets_mean": 5157.9,
      "valid_targets_min": 627
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.20962176596050827,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.3467,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09128950536251068,
      "step": 735,
      "valid_targets_mean": 5631.8,
      "valid_targets_min": 1019
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.2126742923611891,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3478,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08647491037845612,
      "step": 740,
      "valid_targets_mean": 5357.8,
      "valid_targets_min": 719
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.1980996932188349,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.3547,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0895913764834404,
      "step": 745,
      "valid_targets_mean": 6320.7,
      "valid_targets_min": 1813
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.1956495852319613,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.3488,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08508864045143127,
      "step": 750,
      "valid_targets_mean": 5826.7,
      "valid_targets_min": 747
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.20326077871311263,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.3469,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08511250466108322,
      "step": 755,
      "valid_targets_mean": 6488.7,
      "valid_targets_min": 580
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.21531552385542027,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3523,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07421720027923584,
      "step": 760,
      "valid_targets_mean": 4237.1,
      "valid_targets_min": 762
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.209762808261598,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3592,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09930908679962158,
      "step": 765,
      "valid_targets_mean": 5956.4,
      "valid_targets_min": 1315
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.22824690040133508,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.3473,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09878163784742355,
      "step": 770,
      "valid_targets_mean": 5813.5,
      "valid_targets_min": 668
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.20182630283024727,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.3501,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10115042328834534,
      "step": 775,
      "valid_targets_mean": 6430.5,
      "valid_targets_min": 1007
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.20561067266355393,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3637,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0818994864821434,
      "step": 780,
      "valid_targets_mean": 4864.9,
      "valid_targets_min": 580
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.38709523228908527,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3485,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.3410820960998535,
      "step": 785,
      "valid_targets_mean": 5863.9,
      "valid_targets_min": 645
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.3410820960998535,
      "step": 785,
      "total_flos": 1.8466654367420252e+18,
      "train_loss": 0.38092176382708703,
      "train_runtime": 23853.1903,
      "train_samples_per_second": 2.096,
      "train_steps_per_second": 0.033,
      "valid_targets_mean": 5863.9,
      "valid_targets_min": 645
    }
  ],
  "logging_steps": 5,
  "max_steps": 785,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8466654367420252e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}