{
  "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": 7.631121043697976,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.8891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.22271543741226196,
      "step": 5,
      "valid_targets_mean": 2476.6,
      "valid_targets_min": 1301
    },
    {
      "epoch": 0.064,
      "grad_norm": 5.27881420878073,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.8953,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.20072758197784424,
      "step": 10,
      "valid_targets_mean": 2332.7,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.096,
      "grad_norm": 2.6810809161759894,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.8167,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.21659545600414276,
      "step": 15,
      "valid_targets_mean": 2635.2,
      "valid_targets_min": 2017
    },
    {
      "epoch": 0.128,
      "grad_norm": 1.3612409092128206,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.7665,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.20028266310691833,
      "step": 20,
      "valid_targets_mean": 2392.8,
      "valid_targets_min": 834
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.9712808216614238,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.7092,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16015717387199402,
      "step": 25,
      "valid_targets_mean": 2036.3,
      "valid_targets_min": 627
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.7847495702942517,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.6989,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16251808404922485,
      "step": 30,
      "valid_targets_mean": 2170.1,
      "valid_targets_min": 695
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.6368837163308867,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.6551,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.18215426802635193,
      "step": 35,
      "valid_targets_mean": 2561.9,
      "valid_targets_min": 1349
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.4861435916193186,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.6345,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1406915932893753,
      "step": 40,
      "valid_targets_mean": 2587.4,
      "valid_targets_min": 1156
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.42086791320978584,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.5955,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12135272473096848,
      "step": 45,
      "valid_targets_mean": 2122.2,
      "valid_targets_min": 505
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.419828484849555,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.5878,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13206100463867188,
      "step": 50,
      "valid_targets_mean": 2235.4,
      "valid_targets_min": 891
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.3947927235863584,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.5708,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1331617832183838,
      "step": 55,
      "valid_targets_mean": 2493.7,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.3677300729554212,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.5517,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13664039969444275,
      "step": 60,
      "valid_targets_mean": 2516.1,
      "valid_targets_min": 748
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.4072083040661874,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.572,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12455327808856964,
      "step": 65,
      "valid_targets_mean": 2005.4,
      "valid_targets_min": 666
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.3859323974434273,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.5393,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1278195083141327,
      "step": 70,
      "valid_targets_mean": 2248.9,
      "valid_targets_min": 770
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3618871599465855,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.5479,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12145921587944031,
      "step": 75,
      "valid_targets_mean": 2457.3,
      "valid_targets_min": 442
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.3778666430950872,
      "learning_rate": 4e-05,
      "loss": 0.5262,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1243448480963707,
      "step": 80,
      "valid_targets_mean": 2133.5,
      "valid_targets_min": 712
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.36204735116857517,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.5216,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11901817470788956,
      "step": 85,
      "valid_targets_mean": 2211.9,
      "valid_targets_min": 659
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.37136134435893786,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.513,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11105550080537796,
      "step": 90,
      "valid_targets_mean": 1993.4,
      "valid_targets_min": 670
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.37525675194181013,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.5231,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1376522183418274,
      "step": 95,
      "valid_targets_mean": 2684.2,
      "valid_targets_min": 974
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3824297270784445,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.5053,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1333228498697281,
      "step": 100,
      "valid_targets_mean": 2383.8,
      "valid_targets_min": 886
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.37527053538555377,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.5058,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14271236956119537,
      "step": 105,
      "valid_targets_mean": 2695.1,
      "valid_targets_min": 760
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.3537104082629273,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.5003,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12837614119052887,
      "step": 110,
      "valid_targets_mean": 2870.2,
      "valid_targets_min": 1437
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.3509645219531119,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.4992,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12494358420372009,
      "step": 115,
      "valid_targets_mean": 2231.8,
      "valid_targets_min": 576
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.37140995188960296,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.501,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13252729177474976,
      "step": 120,
      "valid_targets_mean": 2579.4,
      "valid_targets_min": 1069
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.38592521508298305,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.5014,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10906288027763367,
      "step": 125,
      "valid_targets_mean": 2436.2,
      "valid_targets_min": 1019
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.3774469495344788,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.5088,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1347123384475708,
      "step": 130,
      "valid_targets_mean": 2156.8,
      "valid_targets_min": 687
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.35331778619634596,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.5004,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11867493391036987,
      "step": 135,
      "valid_targets_mean": 2269.1,
      "valid_targets_min": 496
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.381269763301234,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.4846,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11897442489862442,
      "step": 140,
      "valid_targets_mean": 2282.8,
      "valid_targets_min": 729
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.37635071567491307,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.4762,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12302634119987488,
      "step": 145,
      "valid_targets_mean": 2549.6,
      "valid_targets_min": 812
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.3696178527878047,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.4781,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13179175555706024,
      "step": 150,
      "valid_targets_mean": 2555.1,
      "valid_targets_min": 987
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.3583947059528666,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.4705,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11860240995883942,
      "step": 155,
      "valid_targets_mean": 2559.3,
      "valid_targets_min": 676
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.3621488113478336,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.4765,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11035702377557755,
      "step": 160,
      "valid_targets_mean": 2256.6,
      "valid_targets_min": 575
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.38027772195572573,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.4875,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12132596969604492,
      "step": 165,
      "valid_targets_mean": 2224.2,
      "valid_targets_min": 683
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.3831848460828967,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.4732,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.111229307949543,
      "step": 170,
      "valid_targets_mean": 2282.8,
      "valid_targets_min": 858
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.3882380505987559,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.4665,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11933496594429016,
      "step": 175,
      "valid_targets_mean": 2197.6,
      "valid_targets_min": 535
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.38808772044969597,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.4633,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1027202382683754,
      "step": 180,
      "valid_targets_mean": 2257.2,
      "valid_targets_min": 1237
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.36293154506315095,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.4553,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11203598231077194,
      "step": 185,
      "valid_targets_mean": 2293.1,
      "valid_targets_min": 826
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.3689772416334428,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.4662,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11721555888652802,
      "step": 190,
      "valid_targets_mean": 2506.8,
      "valid_targets_min": 735
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.43777762642017926,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.4618,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12152767926454544,
      "step": 195,
      "valid_targets_mean": 3030.2,
      "valid_targets_min": 1859
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.371935582678312,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.466,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11050638556480408,
      "step": 200,
      "valid_targets_mean": 2149.6,
      "valid_targets_min": 824
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.37190151848447794,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.4619,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10513182729482651,
      "step": 205,
      "valid_targets_mean": 2345.4,
      "valid_targets_min": 712
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.36687001281293485,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.4718,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13081791996955872,
      "step": 210,
      "valid_targets_mean": 3058.9,
      "valid_targets_min": 1310
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.5244127872645981,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.4771,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11074385046958923,
      "step": 215,
      "valid_targets_mean": 2053.8,
      "valid_targets_min": 544
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.37485388401896785,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.4658,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12147005647420883,
      "step": 220,
      "valid_targets_mean": 2652.4,
      "valid_targets_min": 615
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.36922684378704207,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.4672,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12403183430433273,
      "step": 225,
      "valid_targets_mean": 2192.9,
      "valid_targets_min": 747
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.36374441036147753,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.4672,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1292041838169098,
      "step": 230,
      "valid_targets_mean": 3004.4,
      "valid_targets_min": 1832
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.3948554267634894,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.4598,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12679752707481384,
      "step": 235,
      "valid_targets_mean": 2637.1,
      "valid_targets_min": 946
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.37068345902272504,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.4661,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09907030314207077,
      "step": 240,
      "valid_targets_mean": 2072.6,
      "valid_targets_min": 627
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.3669539251059158,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.4641,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11002381145954132,
      "step": 245,
      "valid_targets_mean": 2329.2,
      "valid_targets_min": 1380
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.4096014954540597,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.4703,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11717953532934189,
      "step": 250,
      "valid_targets_mean": 2277.5,
      "valid_targets_min": 988
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.39509938189105553,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.4594,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10379903018474579,
      "step": 255,
      "valid_targets_mean": 2348.1,
      "valid_targets_min": 699
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.3756191077532937,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.4615,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11246967315673828,
      "step": 260,
      "valid_targets_mean": 2493.1,
      "valid_targets_min": 840
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.4044690200794231,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.4649,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.110806405544281,
      "step": 265,
      "valid_targets_mean": 2017.8,
      "valid_targets_min": 717
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.5081640504507091,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.4549,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11942463368177414,
      "step": 270,
      "valid_targets_mean": 2584.5,
      "valid_targets_min": 834
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.3927226443344469,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.4566,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12417890131473541,
      "step": 275,
      "valid_targets_mean": 2363.7,
      "valid_targets_min": 1147
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.3823668344189796,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.4562,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12030941992998123,
      "step": 280,
      "valid_targets_mean": 2514.6,
      "valid_targets_min": 1588
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.4080217126524122,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.4522,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13041387498378754,
      "step": 285,
      "valid_targets_mean": 2502.2,
      "valid_targets_min": 994
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.35720690950041145,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.4543,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09930306673049927,
      "step": 290,
      "valid_targets_mean": 2340.0,
      "valid_targets_min": 672
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.3784197895262166,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.4672,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11923228949308395,
      "step": 295,
      "valid_targets_mean": 2192.8,
      "valid_targets_min": 805
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.3853261461806844,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.4582,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13272637128829956,
      "step": 300,
      "valid_targets_mean": 2641.6,
      "valid_targets_min": 718
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.4086565083973346,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.4568,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11800284683704376,
      "step": 305,
      "valid_targets_mean": 2454.9,
      "valid_targets_min": 605
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.37246762381338155,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.4648,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1008586436510086,
      "step": 310,
      "valid_targets_mean": 1896.5,
      "valid_targets_min": 663
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.3806650012392375,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.4546,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11685250699520111,
      "step": 315,
      "valid_targets_mean": 2345.5,
      "valid_targets_min": 1122
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.3798085186564654,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.4469,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12193480134010315,
      "step": 320,
      "valid_targets_mean": 2326.6,
      "valid_targets_min": 842
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.36905791747982847,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.4457,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1169414296746254,
      "step": 325,
      "valid_targets_mean": 2857.9,
      "valid_targets_min": 962
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.3402606779146125,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.4362,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09596966952085495,
      "step": 330,
      "valid_targets_mean": 2375.1,
      "valid_targets_min": 631
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.36648780234004474,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.4407,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10836751013994217,
      "step": 335,
      "valid_targets_mean": 2527.1,
      "valid_targets_min": 831
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.39668264319464347,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.4459,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12774476408958435,
      "step": 340,
      "valid_targets_mean": 2352.4,
      "valid_targets_min": 798
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.36386446276091866,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.4454,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10319218039512634,
      "step": 345,
      "valid_targets_mean": 2521.3,
      "valid_targets_min": 1315
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.3546475260205052,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10272243618965149,
      "step": 350,
      "valid_targets_mean": 2621.2,
      "valid_targets_min": 773
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.3630552906576787,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.4325,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10560596734285355,
      "step": 355,
      "valid_targets_mean": 2550.9,
      "valid_targets_min": 779
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.38279271826493005,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.4351,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1132790818810463,
      "step": 360,
      "valid_targets_mean": 2419.3,
      "valid_targets_min": 720
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.39935145796247956,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.4414,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09891170263290405,
      "step": 365,
      "valid_targets_mean": 2249.1,
      "valid_targets_min": 1164
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.3419394392882838,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.4368,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10849423706531525,
      "step": 370,
      "valid_targets_mean": 2371.8,
      "valid_targets_min": 843
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.3771680016371104,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.4442,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1079641729593277,
      "step": 375,
      "valid_targets_mean": 2512.6,
      "valid_targets_min": 475
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.3707439432584615,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.4433,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1102166473865509,
      "step": 380,
      "valid_targets_mean": 2372.2,
      "valid_targets_min": 724
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.3688167243996532,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.4379,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12143057584762573,
      "step": 385,
      "valid_targets_mean": 2571.7,
      "valid_targets_min": 1490
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.372498854103246,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.45,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10243070125579834,
      "step": 390,
      "valid_targets_mean": 2157.3,
      "valid_targets_min": 603
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.3561390151542751,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.4435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11096415668725967,
      "step": 395,
      "valid_targets_mean": 2564.1,
      "valid_targets_min": 1146
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.3467986057670459,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.4391,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09745585918426514,
      "step": 400,
      "valid_targets_mean": 2261.1,
      "valid_targets_min": 484
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.38784817744913735,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.4456,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10883487015962601,
      "step": 405,
      "valid_targets_mean": 2358.3,
      "valid_targets_min": 589
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.39520877797364024,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.4491,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11191672086715698,
      "step": 410,
      "valid_targets_mean": 2358.7,
      "valid_targets_min": 861
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.3501064725751044,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.4376,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10743163526058197,
      "step": 415,
      "valid_targets_mean": 2539.9,
      "valid_targets_min": 615
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.34747271508177885,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.4406,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11140751093626022,
      "step": 420,
      "valid_targets_mean": 2474.9,
      "valid_targets_min": 644
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.32707554036034925,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.4396,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11559329926967621,
      "step": 425,
      "valid_targets_mean": 3053.6,
      "valid_targets_min": 920
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.38640280327503035,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.4416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12284263968467712,
      "step": 430,
      "valid_targets_mean": 2602.9,
      "valid_targets_min": 585
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.4114312862720891,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.4334,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11041560769081116,
      "step": 435,
      "valid_targets_mean": 2347.3,
      "valid_targets_min": 619
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.32073194733072063,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.4356,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11413581669330597,
      "step": 440,
      "valid_targets_mean": 3212.8,
      "valid_targets_min": 997
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.46398697838839303,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.4403,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10393176227807999,
      "step": 445,
      "valid_targets_mean": 2268.5,
      "valid_targets_min": 972
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.36496198308115707,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.4333,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10166573524475098,
      "step": 450,
      "valid_targets_mean": 2378.6,
      "valid_targets_min": 535
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.3568647386974551,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.4366,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1209319531917572,
      "step": 455,
      "valid_targets_mean": 2209.4,
      "valid_targets_min": 956
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.36074018847887357,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.4425,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1085042655467987,
      "step": 460,
      "valid_targets_mean": 2346.1,
      "valid_targets_min": 494
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.3543561136932817,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.4351,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10610386729240417,
      "step": 465,
      "valid_targets_mean": 2401.0,
      "valid_targets_min": 783
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.353857442044966,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1253969669342041,
      "step": 470,
      "valid_targets_mean": 2643.6,
      "valid_targets_min": 924
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.3510915710318982,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.4272,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1131058931350708,
      "step": 475,
      "valid_targets_mean": 3003.8,
      "valid_targets_min": 1878
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.3834167334438327,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.4323,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11351168155670166,
      "step": 480,
      "valid_targets_mean": 2438.5,
      "valid_targets_min": 597
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.35598468413223955,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.4272,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1119239330291748,
      "step": 485,
      "valid_targets_mean": 2589.8,
      "valid_targets_min": 839
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.35985279567171535,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.4258,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09996067732572556,
      "step": 490,
      "valid_targets_mean": 2321.4,
      "valid_targets_min": 560
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.37201568967503795,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.4306,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10631603747606277,
      "step": 495,
      "valid_targets_mean": 2443.0,
      "valid_targets_min": 926
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.3683426934856062,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.4293,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10557325184345245,
      "step": 500,
      "valid_targets_mean": 2416.2,
      "valid_targets_min": 761
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.3424560285954452,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.4305,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11586504429578781,
      "step": 505,
      "valid_targets_mean": 2864.2,
      "valid_targets_min": 610
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.3578344128596558,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.4211,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09824161231517792,
      "step": 510,
      "valid_targets_mean": 2193.2,
      "valid_targets_min": 580
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.3561114928124899,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.4297,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10587438941001892,
      "step": 515,
      "valid_targets_mean": 2400.8,
      "valid_targets_min": 724
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.35873753478305465,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.4304,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12063280493021011,
      "step": 520,
      "valid_targets_mean": 2813.1,
      "valid_targets_min": 1160
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.3765778341929378,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.4161,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09327021986246109,
      "step": 525,
      "valid_targets_mean": 1924.0,
      "valid_targets_min": 617
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.4324457664668057,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.4172,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10706986486911774,
      "step": 530,
      "valid_targets_mean": 2922.2,
      "valid_targets_min": 607
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.36145748070452655,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.4311,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11657077074050903,
      "step": 535,
      "valid_targets_mean": 2529.0,
      "valid_targets_min": 760
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.34484082874160493,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.4354,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.105072021484375,
      "step": 540,
      "valid_targets_mean": 2642.2,
      "valid_targets_min": 855
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.3516326667725607,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.4215,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1013285294175148,
      "step": 545,
      "valid_targets_mean": 2099.8,
      "valid_targets_min": 631
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.37571166625529,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.4324,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11065156757831573,
      "step": 550,
      "valid_targets_mean": 2494.1,
      "valid_targets_min": 1062
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.34979910368925665,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.4321,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09944584965705872,
      "step": 555,
      "valid_targets_mean": 2163.4,
      "valid_targets_min": 787
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.3572748730592331,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.434,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11576366424560547,
      "step": 560,
      "valid_targets_mean": 2483.0,
      "valid_targets_min": 544
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.35967280637810656,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.4292,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10254612565040588,
      "step": 565,
      "valid_targets_mean": 2525.1,
      "valid_targets_min": 762
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.3455425957064251,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.4277,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10262332856655121,
      "step": 570,
      "valid_targets_mean": 2300.0,
      "valid_targets_min": 627
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.3688641590124713,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.4289,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13081635534763336,
      "step": 575,
      "valid_targets_mean": 2898.2,
      "valid_targets_min": 1551
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.35149511407843115,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.4279,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10150306671857834,
      "step": 580,
      "valid_targets_mean": 2345.7,
      "valid_targets_min": 965
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.37010020670114396,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.4309,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12104976922273636,
      "step": 585,
      "valid_targets_mean": 2471.3,
      "valid_targets_min": 748
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.329279647087639,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.4247,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09903521835803986,
      "step": 590,
      "valid_targets_mean": 2431.8,
      "valid_targets_min": 949
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.3706881450643897,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.4206,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10147304832935333,
      "step": 595,
      "valid_targets_mean": 2359.5,
      "valid_targets_min": 1145
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.3544528487803963,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.4212,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09534944593906403,
      "step": 600,
      "valid_targets_mean": 2434.1,
      "valid_targets_min": 673
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.34572960602120706,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.424,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10098496079444885,
      "step": 605,
      "valid_targets_mean": 2049.7,
      "valid_targets_min": 672
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.3385251496774486,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.4309,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11481376737356186,
      "step": 610,
      "valid_targets_mean": 2369.4,
      "valid_targets_min": 633
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.6228350973720654,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.4262,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11105981469154358,
      "step": 615,
      "valid_targets_mean": 2713.6,
      "valid_targets_min": 841
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.4755804075607999,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.4315,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1181117370724678,
      "step": 620,
      "valid_targets_mean": 2359.4,
      "valid_targets_min": 1621
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.35327709484906245,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.4264,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10537832975387573,
      "step": 625,
      "valid_targets_mean": 2640.1,
      "valid_targets_min": 996
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.366071511572943,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.4178,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11115087568759918,
      "step": 630,
      "valid_targets_mean": 2280.6,
      "valid_targets_min": 1098
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.3746937074370881,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.4274,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10405690968036652,
      "step": 635,
      "valid_targets_mean": 1970.4,
      "valid_targets_min": 610
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.33933207284418376,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.4183,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10596753656864166,
      "step": 640,
      "valid_targets_mean": 2338.5,
      "valid_targets_min": 1075
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.36652871749180194,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.4245,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10570182651281357,
      "step": 645,
      "valid_targets_mean": 2398.5,
      "valid_targets_min": 679
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.34666144785143826,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.4223,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09318490326404572,
      "step": 650,
      "valid_targets_mean": 2170.8,
      "valid_targets_min": 524
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.349600144940352,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.4122,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10870470106601715,
      "step": 655,
      "valid_targets_mean": 2790.6,
      "valid_targets_min": 1832
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.3657293355482953,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.4263,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11297493427991867,
      "step": 660,
      "valid_targets_mean": 2327.8,
      "valid_targets_min": 900
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.33512568991595465,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.4204,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09641289710998535,
      "step": 665,
      "valid_targets_mean": 2293.8,
      "valid_targets_min": 689
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.3361973926134737,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.4192,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08825628459453583,
      "step": 670,
      "valid_targets_mean": 2078.7,
      "valid_targets_min": 633
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.40275466302733576,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.4059,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11168158054351807,
      "step": 675,
      "valid_targets_mean": 2825.6,
      "valid_targets_min": 814
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.33881094238458337,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.4212,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09750144183635712,
      "step": 680,
      "valid_targets_mean": 2606.6,
      "valid_targets_min": 667
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.34649225347244983,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.4283,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09878174960613251,
      "step": 685,
      "valid_targets_mean": 2184.4,
      "valid_targets_min": 810
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.6416853579629246,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.4209,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10486841201782227,
      "step": 690,
      "valid_targets_mean": 2410.8,
      "valid_targets_min": 562
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.34242289858813213,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.4163,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11274117976427078,
      "step": 695,
      "valid_targets_mean": 2641.4,
      "valid_targets_min": 697
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.4489652104488755,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.4288,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10218752920627594,
      "step": 700,
      "valid_targets_mean": 2205.9,
      "valid_targets_min": 693
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.38750863503880256,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.4224,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11015324294567108,
      "step": 705,
      "valid_targets_mean": 2522.4,
      "valid_targets_min": 989
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.3638004944669233,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.4219,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11784189939498901,
      "step": 710,
      "valid_targets_mean": 2423.0,
      "valid_targets_min": 832
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.33854575338443277,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.4222,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11596603691577911,
      "step": 715,
      "valid_targets_mean": 2593.8,
      "valid_targets_min": 960
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.3426098834958257,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.4227,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10185693204402924,
      "step": 720,
      "valid_targets_mean": 2219.6,
      "valid_targets_min": 1182
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.35018898874669724,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.4279,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12424006313085556,
      "step": 725,
      "valid_targets_mean": 2555.5,
      "valid_targets_min": 1024
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.32903025513626405,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.4177,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09462594985961914,
      "step": 730,
      "valid_targets_mean": 2214.7,
      "valid_targets_min": 627
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.355942373474534,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.4242,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1105259582400322,
      "step": 735,
      "valid_targets_mean": 2385.1,
      "valid_targets_min": 1019
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.3440268419923915,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.4208,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09921759366989136,
      "step": 740,
      "valid_targets_mean": 2268.0,
      "valid_targets_min": 719
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.3391292108191765,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.423,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10051105916500092,
      "step": 745,
      "valid_targets_mean": 2327.6,
      "valid_targets_min": 1502
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.34166487931018136,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.4167,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10346902906894684,
      "step": 750,
      "valid_targets_mean": 2279.1,
      "valid_targets_min": 747
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.3214774241350311,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.4164,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10058611631393433,
      "step": 755,
      "valid_targets_mean": 3048.1,
      "valid_targets_min": 580
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.3387603005789863,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.4288,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09350170195102692,
      "step": 760,
      "valid_targets_mean": 2165.1,
      "valid_targets_min": 762
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.33445320680299534,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.4178,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10848960280418396,
      "step": 765,
      "valid_targets_mean": 2675.0,
      "valid_targets_min": 1314
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.3423233493479509,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.4176,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10347608476877213,
      "step": 770,
      "valid_targets_mean": 2338.2,
      "valid_targets_min": 668
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.3242922001719799,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.4207,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11179254204034805,
      "step": 775,
      "valid_targets_mean": 2694.2,
      "valid_targets_min": 1007
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.33364671731962425,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.4281,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09912078082561493,
      "step": 780,
      "valid_targets_mean": 2130.0,
      "valid_targets_min": 580
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.7100559447088324,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.4208,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4209003150463104,
      "step": 785,
      "valid_targets_mean": 2253.5,
      "valid_targets_min": 645
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4209003150463104,
      "step": 785,
      "total_flos": 8.136605209670451e+17,
      "train_loss": 0.4667121115763476,
      "train_runtime": 9534.5841,
      "train_samples_per_second": 5.242,
      "train_steps_per_second": 0.082,
      "valid_targets_mean": 2253.5,
      "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": 8.136605209670451e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}