{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 785,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032,
      "grad_norm": 6.048756324166434,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.7749,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1723039150238037,
      "step": 5,
      "valid_targets_mean": 3899.1,
      "valid_targets_min": 639
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.820317754564466,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.7446,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.185373917222023,
      "step": 10,
      "valid_targets_mean": 5002.5,
      "valid_targets_min": 1422
    },
    {
      "epoch": 0.096,
      "grad_norm": 2.5037828274167304,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.6931,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17158135771751404,
      "step": 15,
      "valid_targets_mean": 3496.6,
      "valid_targets_min": 1215
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.9315157784215937,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.6545,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17365200817584991,
      "step": 20,
      "valid_targets_mean": 4261.8,
      "valid_targets_min": 1995
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7754812519149545,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.611,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12669777870178223,
      "step": 25,
      "valid_targets_mean": 2931.9,
      "valid_targets_min": 566
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.52185829607096,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.5995,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14718282222747803,
      "step": 30,
      "valid_targets_mean": 4264.7,
      "valid_targets_min": 1204
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.4852281885120764,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5804,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13390770554542542,
      "step": 35,
      "valid_targets_mean": 3495.6,
      "valid_targets_min": 600
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.38657633820848264,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.5536,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11915287375450134,
      "step": 40,
      "valid_targets_mean": 4156.8,
      "valid_targets_min": 816
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.3612407081777249,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.5628,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15402325987815857,
      "step": 45,
      "valid_targets_mean": 4639.7,
      "valid_targets_min": 982
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.32318118714364474,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.522,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13298863172531128,
      "step": 50,
      "valid_targets_mean": 3814.6,
      "valid_targets_min": 655
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.29866719779423245,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.4995,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12222938239574432,
      "step": 55,
      "valid_targets_mean": 3945.7,
      "valid_targets_min": 791
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.30726596921009003,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.4916,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11845584958791733,
      "step": 60,
      "valid_targets_mean": 3824.1,
      "valid_targets_min": 755
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.2521297207306386,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.4989,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1335579752922058,
      "step": 65,
      "valid_targets_mean": 4825.6,
      "valid_targets_min": 2415
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.26719933666277185,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.4844,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12163548916578293,
      "step": 70,
      "valid_targets_mean": 4284.9,
      "valid_targets_min": 535
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2979141750380719,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.4583,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11367259174585342,
      "step": 75,
      "valid_targets_mean": 4516.4,
      "valid_targets_min": 911
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.30292473042949297,
      "learning_rate": 4e-05,
      "loss": 0.4716,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12441614270210266,
      "step": 80,
      "valid_targets_mean": 4018.8,
      "valid_targets_min": 707
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.28266530877838175,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.4824,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10960252583026886,
      "step": 85,
      "valid_targets_mean": 3031.5,
      "valid_targets_min": 902
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.287898903894681,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.4586,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12198155373334885,
      "step": 90,
      "valid_targets_mean": 4651.1,
      "valid_targets_min": 1158
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.2696418518664897,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4502,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1148751974105835,
      "step": 95,
      "valid_targets_mean": 4679.6,
      "valid_targets_min": 485
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.29505853573744345,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.4575,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10742583870887756,
      "step": 100,
      "valid_targets_mean": 3360.1,
      "valid_targets_min": 697
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.2941887926453614,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.4541,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10949065536260605,
      "step": 105,
      "valid_targets_mean": 4291.4,
      "valid_targets_min": 853
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.30772503562874043,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4564,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11388377845287323,
      "step": 110,
      "valid_targets_mean": 3922.4,
      "valid_targets_min": 586
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.2894203299726909,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.441,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1191316694021225,
      "step": 115,
      "valid_targets_mean": 5007.5,
      "valid_targets_min": 895
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.3360267509201506,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.4523,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11252386122941971,
      "step": 120,
      "valid_targets_mean": 3807.6,
      "valid_targets_min": 783
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2954655902404845,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4562,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10796886682510376,
      "step": 125,
      "valid_targets_mean": 3645.1,
      "valid_targets_min": 1120
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.2979238438193785,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.4303,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1278803050518036,
      "step": 130,
      "valid_targets_mean": 4754.6,
      "valid_targets_min": 826
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.3025824931420387,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.4324,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11299727112054825,
      "step": 135,
      "valid_targets_mean": 3936.9,
      "valid_targets_min": 651
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.27361938948965725,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.4393,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11303415894508362,
      "step": 140,
      "valid_targets_mean": 4267.7,
      "valid_targets_min": 674
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.26468614400279594,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.4197,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1154870092868805,
      "step": 145,
      "valid_targets_mean": 4747.6,
      "valid_targets_min": 1718
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.4577347781033962,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.4496,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12438996136188507,
      "step": 150,
      "valid_targets_mean": 4225.4,
      "valid_targets_min": 906
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.28366351389299993,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.4352,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10910991579294205,
      "step": 155,
      "valid_targets_mean": 4543.1,
      "valid_targets_min": 723
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.2851059883525269,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.4226,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09767496585845947,
      "step": 160,
      "valid_targets_mean": 3813.6,
      "valid_targets_min": 739
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.27633015571631864,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.4195,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09590106457471848,
      "step": 165,
      "valid_targets_mean": 4107.2,
      "valid_targets_min": 851
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.29862294491293667,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.4365,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11445736140012741,
      "step": 170,
      "valid_targets_mean": 4095.9,
      "valid_targets_min": 1214
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.2665048493087847,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.4221,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10399468243122101,
      "step": 175,
      "valid_targets_mean": 4598.8,
      "valid_targets_min": 1231
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.3020735192506363,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.413,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11090105026960373,
      "step": 180,
      "valid_targets_mean": 4057.2,
      "valid_targets_min": 887
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.27625874894849545,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.4155,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09643165022134781,
      "step": 185,
      "valid_targets_mean": 4063.6,
      "valid_targets_min": 988
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.3022623050650015,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.4387,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11312417685985565,
      "step": 190,
      "valid_targets_mean": 3933.4,
      "valid_targets_min": 877
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.3109759147204714,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.4174,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11555179208517075,
      "step": 195,
      "valid_targets_mean": 3763.6,
      "valid_targets_min": 995
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.2923130244414311,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.4229,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11472899466753006,
      "step": 200,
      "valid_targets_mean": 4125.8,
      "valid_targets_min": 938
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.2782752287912167,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.4211,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10930650681257248,
      "step": 205,
      "valid_targets_mean": 4209.9,
      "valid_targets_min": 542
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.2991828057882393,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10878201574087143,
      "step": 210,
      "valid_targets_mean": 4023.2,
      "valid_targets_min": 773
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.30997637082473456,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.4162,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08643242716789246,
      "step": 215,
      "valid_targets_mean": 3282.3,
      "valid_targets_min": 761
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.2715643458114872,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.4217,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10085868835449219,
      "step": 220,
      "valid_targets_mean": 3737.1,
      "valid_targets_min": 661
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.3420125787516997,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.4234,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11982588469982147,
      "step": 225,
      "valid_targets_mean": 4721.4,
      "valid_targets_min": 559
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.3500123406955092,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.4166,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09180434793233871,
      "step": 230,
      "valid_targets_mean": 4225.7,
      "valid_targets_min": 640
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.27575481164813326,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.4274,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09477457404136658,
      "step": 235,
      "valid_targets_mean": 3827.4,
      "valid_targets_min": 579
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.308272520000558,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.4098,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1010490283370018,
      "step": 240,
      "valid_targets_mean": 4072.6,
      "valid_targets_min": 875
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.28723889701430083,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.4086,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10771428048610687,
      "step": 245,
      "valid_targets_mean": 5469.1,
      "valid_targets_min": 816
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.31601519650893833,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.4132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11569797992706299,
      "step": 250,
      "valid_targets_mean": 4334.8,
      "valid_targets_min": 688
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.35135975450130674,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.4265,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1070767492055893,
      "step": 255,
      "valid_targets_mean": 3814.1,
      "valid_targets_min": 785
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.29723308168544715,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.4188,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10516910254955292,
      "step": 260,
      "valid_targets_mean": 3706.8,
      "valid_targets_min": 800
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.2768385567348046,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.4055,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09208667278289795,
      "step": 265,
      "valid_targets_mean": 3703.6,
      "valid_targets_min": 550
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.2951856918024355,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.4227,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1036691889166832,
      "step": 270,
      "valid_targets_mean": 3547.4,
      "valid_targets_min": 713
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.26249382923658837,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.4182,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09299944341182709,
      "step": 275,
      "valid_targets_mean": 4366.1,
      "valid_targets_min": 794
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.29210022832805477,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.3992,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10157176852226257,
      "step": 280,
      "valid_targets_mean": 3761.3,
      "valid_targets_min": 696
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.27904682672912556,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.4214,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09509153664112091,
      "step": 285,
      "valid_targets_mean": 3494.4,
      "valid_targets_min": 582
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.3252762284277571,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.4189,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1019916832447052,
      "step": 290,
      "valid_targets_mean": 3652.1,
      "valid_targets_min": 851
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.2559216215832805,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.4036,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08774925768375397,
      "step": 295,
      "valid_targets_mean": 3680.8,
      "valid_targets_min": 785
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.28129794947346765,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.4012,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10966770350933075,
      "step": 300,
      "valid_targets_mean": 4628.2,
      "valid_targets_min": 585
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.3261593902401628,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.4036,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10367356240749359,
      "step": 305,
      "valid_targets_mean": 4310.1,
      "valid_targets_min": 892
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.2946580636329257,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.4056,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10338163375854492,
      "step": 310,
      "valid_targets_mean": 4015.2,
      "valid_targets_min": 934
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.3039500754589427,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.4175,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09754990041255951,
      "step": 315,
      "valid_targets_mean": 3799.8,
      "valid_targets_min": 754
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.295008199316107,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.3919,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09975887835025787,
      "step": 320,
      "valid_targets_mean": 3814.1,
      "valid_targets_min": 654
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.28796119211965765,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.4054,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09366752952337265,
      "step": 325,
      "valid_targets_mean": 4051.3,
      "valid_targets_min": 627
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.32513160009341713,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.3913,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10308220982551575,
      "step": 330,
      "valid_targets_mean": 4733.2,
      "valid_targets_min": 1213
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.24307889326670307,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.4093,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08254360407590866,
      "step": 335,
      "valid_targets_mean": 4415.9,
      "valid_targets_min": 549
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.28397293492211034,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.4046,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11996494978666306,
      "step": 340,
      "valid_targets_mean": 4177.9,
      "valid_targets_min": 578
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.28612539051013547,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.4116,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1003495305776596,
      "step": 345,
      "valid_targets_mean": 3604.4,
      "valid_targets_min": 627
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.2901012203725845,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.4047,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10313699394464493,
      "step": 350,
      "valid_targets_mean": 4313.1,
      "valid_targets_min": 830
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.339340824279522,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.4084,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09717093408107758,
      "step": 355,
      "valid_targets_mean": 2634.5,
      "valid_targets_min": 622
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.3356431583635563,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.3965,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11940605938434601,
      "step": 360,
      "valid_targets_mean": 4358.6,
      "valid_targets_min": 653
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.2729074469449499,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.3957,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10392723232507706,
      "step": 365,
      "valid_targets_mean": 4498.1,
      "valid_targets_min": 1129
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.2893023850029988,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.3944,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11088322103023529,
      "step": 370,
      "valid_targets_mean": 4274.2,
      "valid_targets_min": 593
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.25313367295678246,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3795,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09681592136621475,
      "step": 375,
      "valid_targets_mean": 4609.8,
      "valid_targets_min": 799
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.2936301548074808,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.3992,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10911795496940613,
      "step": 380,
      "valid_targets_mean": 3933.9,
      "valid_targets_min": 962
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.2731075417593816,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.4039,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11098127067089081,
      "step": 385,
      "valid_targets_mean": 4835.6,
      "valid_targets_min": 587
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.291325200016876,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.4048,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09697733074426651,
      "step": 390,
      "valid_targets_mean": 3457.2,
      "valid_targets_min": 803
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.28348123924932705,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.3915,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09868316352367401,
      "step": 395,
      "valid_targets_mean": 4449.7,
      "valid_targets_min": 571
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.26609595339552466,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.3895,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0807412639260292,
      "step": 400,
      "valid_targets_mean": 3627.1,
      "valid_targets_min": 661
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.33358859230174703,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.398,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10890471935272217,
      "step": 405,
      "valid_targets_mean": 3517.1,
      "valid_targets_min": 585
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.3225642644827852,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.3952,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10589894652366638,
      "step": 410,
      "valid_targets_mean": 3461.2,
      "valid_targets_min": 483
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.28098105618711433,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.3962,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09055127203464508,
      "step": 415,
      "valid_targets_mean": 3736.4,
      "valid_targets_min": 919
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.28850076772839844,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.3939,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10580737888813019,
      "step": 420,
      "valid_targets_mean": 3956.4,
      "valid_targets_min": 840
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.2640927061351806,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.4044,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10874482244253159,
      "step": 425,
      "valid_targets_mean": 4151.2,
      "valid_targets_min": 734
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.2870930699056917,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.3872,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08751390129327774,
      "step": 430,
      "valid_targets_mean": 3402.9,
      "valid_targets_min": 658
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.2676262206004521,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.3992,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10693050920963287,
      "step": 435,
      "valid_targets_mean": 4426.9,
      "valid_targets_min": 1543
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.2717783395966012,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3901,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10182879120111465,
      "step": 440,
      "valid_targets_mean": 4275.3,
      "valid_targets_min": 720
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.28314115940486834,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.4143,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09646420925855637,
      "step": 445,
      "valid_targets_mean": 4359.9,
      "valid_targets_min": 585
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.27804663750003045,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3923,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10060509294271469,
      "step": 450,
      "valid_targets_mean": 3812.6,
      "valid_targets_min": 446
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.28764089042619995,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.4051,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10025319457054138,
      "step": 455,
      "valid_targets_mean": 3917.0,
      "valid_targets_min": 558
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.2505864627102052,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.3932,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09689773619174957,
      "step": 460,
      "valid_targets_mean": 4338.6,
      "valid_targets_min": 1069
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.23037767347131669,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.3841,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09673911333084106,
      "step": 465,
      "valid_targets_mean": 4582.5,
      "valid_targets_min": 810
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.27508918041338687,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.3979,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09660154581069946,
      "step": 470,
      "valid_targets_mean": 3506.7,
      "valid_targets_min": 618
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.278832581896705,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.3801,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09370477497577667,
      "step": 475,
      "valid_targets_mean": 3882.2,
      "valid_targets_min": 999
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.27988835500157977,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.3895,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1039055809378624,
      "step": 480,
      "valid_targets_mean": 3686.8,
      "valid_targets_min": 767
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.27300541572988496,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.392,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11811760067939758,
      "step": 485,
      "valid_targets_mean": 5150.7,
      "valid_targets_min": 1112
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.2898076648663478,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.3881,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11502294987440109,
      "step": 490,
      "valid_targets_mean": 4739.1,
      "valid_targets_min": 1357
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.2820293732275994,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.3796,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09026939421892166,
      "step": 495,
      "valid_targets_mean": 3438.8,
      "valid_targets_min": 613
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.2563623556654072,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.3891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07436763495206833,
      "step": 500,
      "valid_targets_mean": 3468.8,
      "valid_targets_min": 626
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.2639018773325639,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3872,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0847480371594429,
      "step": 505,
      "valid_targets_mean": 3150.1,
      "valid_targets_min": 720
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.28414234275420414,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3972,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11753346771001816,
      "step": 510,
      "valid_targets_mean": 4629.7,
      "valid_targets_min": 1059
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.24480188835818936,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.3837,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09535688906908035,
      "step": 515,
      "valid_targets_mean": 4956.4,
      "valid_targets_min": 724
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.27121801267241674,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3931,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10601663589477539,
      "step": 520,
      "valid_targets_mean": 4022.4,
      "valid_targets_min": 1571
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.2728948280882927,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.3947,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10746031999588013,
      "step": 525,
      "valid_targets_mean": 3813.9,
      "valid_targets_min": 887
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.28856355078117213,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3869,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11251718550920486,
      "step": 530,
      "valid_targets_mean": 3836.1,
      "valid_targets_min": 1244
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.2660317815137939,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.384,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0875939428806305,
      "step": 535,
      "valid_targets_mean": 3313.4,
      "valid_targets_min": 923
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.3266588391926982,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.3959,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1070742979645729,
      "step": 540,
      "valid_targets_mean": 4143.3,
      "valid_targets_min": 1343
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.25227560229994916,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.3735,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09336927533149719,
      "step": 545,
      "valid_targets_mean": 4148.4,
      "valid_targets_min": 627
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.29732898476146347,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.3873,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09397479891777039,
      "step": 550,
      "valid_targets_mean": 3452.2,
      "valid_targets_min": 635
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.28103656307884795,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3916,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0950830802321434,
      "step": 555,
      "valid_targets_mean": 3525.2,
      "valid_targets_min": 874
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.27515789289270587,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.3873,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08731421828269958,
      "step": 560,
      "valid_targets_mean": 3457.8,
      "valid_targets_min": 816
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.26595925071142734,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.3903,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0943126454949379,
      "step": 565,
      "valid_targets_mean": 3755.8,
      "valid_targets_min": 916
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.26908934868333345,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3881,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11741252988576889,
      "step": 570,
      "valid_targets_mean": 4463.9,
      "valid_targets_min": 1490
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.26163071888829387,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.3804,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07929997891187668,
      "step": 575,
      "valid_targets_mean": 2767.7,
      "valid_targets_min": 668
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.2762268133622116,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.3876,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08485132455825806,
      "step": 580,
      "valid_targets_mean": 3037.7,
      "valid_targets_min": 726
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.2674780145990845,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.3872,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09651274234056473,
      "step": 585,
      "valid_targets_mean": 4118.8,
      "valid_targets_min": 1165
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.24858944575790964,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3843,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09582122415304184,
      "step": 590,
      "valid_targets_mean": 4209.6,
      "valid_targets_min": 790
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.26046610758966066,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.3849,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08760522305965424,
      "step": 595,
      "valid_targets_mean": 3850.9,
      "valid_targets_min": 844
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.2517690965363384,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.3862,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10714523494243622,
      "step": 600,
      "valid_targets_mean": 5318.4,
      "valid_targets_min": 1305
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.27076985537219617,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3801,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09001049399375916,
      "step": 605,
      "valid_targets_mean": 3394.0,
      "valid_targets_min": 991
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.2825297290428291,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.398,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09313380718231201,
      "step": 610,
      "valid_targets_mean": 3341.5,
      "valid_targets_min": 788
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.26545079772678515,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.3819,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08457700908184052,
      "step": 615,
      "valid_targets_mean": 3106.5,
      "valid_targets_min": 660
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.2567709880978487,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.3847,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08885794878005981,
      "step": 620,
      "valid_targets_mean": 4086.2,
      "valid_targets_min": 743
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.24894909802272616,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.3895,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0957787036895752,
      "step": 625,
      "valid_targets_mean": 4743.1,
      "valid_targets_min": 934
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.2730521094355628,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3959,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09598704427480698,
      "step": 630,
      "valid_targets_mean": 3783.9,
      "valid_targets_min": 1002
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.2517076111149134,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.375,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10451523959636688,
      "step": 635,
      "valid_targets_mean": 4389.2,
      "valid_targets_min": 764
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.26123311146967465,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.3842,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.085345059633255,
      "step": 640,
      "valid_targets_mean": 3496.9,
      "valid_targets_min": 1141
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.28008604571769974,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.3849,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10877778381109238,
      "step": 645,
      "valid_targets_mean": 3883.4,
      "valid_targets_min": 836
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.2810668056658052,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.3942,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.103969506919384,
      "step": 650,
      "valid_targets_mean": 4210.5,
      "valid_targets_min": 1079
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.2679325409425069,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3697,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10508221387863159,
      "step": 655,
      "valid_targets_mean": 3703.6,
      "valid_targets_min": 1064
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.2677720958501848,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.3743,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09485597908496857,
      "step": 660,
      "valid_targets_mean": 3589.9,
      "valid_targets_min": 752
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.25540762686442,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.3828,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08198321610689163,
      "step": 665,
      "valid_targets_mean": 3685.8,
      "valid_targets_min": 587
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.2538363359478502,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3829,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10491339862346649,
      "step": 670,
      "valid_targets_mean": 4984.3,
      "valid_targets_min": 698
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.24929527911049837,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3805,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08019746094942093,
      "step": 675,
      "valid_targets_mean": 3777.9,
      "valid_targets_min": 680
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.2688823154765233,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3892,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11593001335859299,
      "step": 680,
      "valid_targets_mean": 4966.8,
      "valid_targets_min": 1682
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.2681893697414637,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.3824,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09049779176712036,
      "step": 685,
      "valid_targets_mean": 3786.8,
      "valid_targets_min": 553
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.25803539534975506,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3822,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09479928016662598,
      "step": 690,
      "valid_targets_mean": 4407.2,
      "valid_targets_min": 701
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.2721781700256369,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.3817,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09600777924060822,
      "step": 695,
      "valid_targets_mean": 4190.6,
      "valid_targets_min": 707
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.27203666956952044,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.3936,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1137179434299469,
      "step": 700,
      "valid_targets_mean": 4398.9,
      "valid_targets_min": 2071
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.28700859136150275,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3863,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0846642404794693,
      "step": 705,
      "valid_targets_mean": 3038.7,
      "valid_targets_min": 724
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.2529565401900442,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.377,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10368967056274414,
      "step": 710,
      "valid_targets_mean": 4356.4,
      "valid_targets_min": 904
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.2500059167539887,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.3834,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08508186787366867,
      "step": 715,
      "valid_targets_mean": 3736.4,
      "valid_targets_min": 788
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.27361014801466704,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3909,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09339834749698639,
      "step": 720,
      "valid_targets_mean": 3070.9,
      "valid_targets_min": 523
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.25039458600858977,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.3735,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07535479962825775,
      "step": 725,
      "valid_targets_mean": 3382.6,
      "valid_targets_min": 794
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.2846141316902581,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.3891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08932908624410629,
      "step": 730,
      "valid_targets_mean": 3472.3,
      "valid_targets_min": 806
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.25547731029216575,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.3894,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09996099770069122,
      "step": 735,
      "valid_targets_mean": 4633.1,
      "valid_targets_min": 1030
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.25171677147508753,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3803,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08795123547315598,
      "step": 740,
      "valid_targets_mean": 3785.0,
      "valid_targets_min": 545
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.24110208408904205,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.39,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10249342024326324,
      "step": 745,
      "valid_targets_mean": 4599.1,
      "valid_targets_min": 1465
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.272490256449447,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.3892,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09520266950130463,
      "step": 750,
      "valid_targets_mean": 3378.0,
      "valid_targets_min": 551
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.27541100522754647,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.389,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09170030802488327,
      "step": 755,
      "valid_targets_mean": 4346.1,
      "valid_targets_min": 1001
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.24879985587571116,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3775,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08432689309120178,
      "step": 760,
      "valid_targets_mean": 3557.9,
      "valid_targets_min": 481
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.25501966207119003,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3674,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09955944120883942,
      "step": 765,
      "valid_targets_mean": 4503.8,
      "valid_targets_min": 1642
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.24823600991634784,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.3811,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08799436688423157,
      "step": 770,
      "valid_targets_mean": 4153.3,
      "valid_targets_min": 703
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.2925007672229399,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.3864,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10177664458751678,
      "step": 775,
      "valid_targets_mean": 4148.4,
      "valid_targets_min": 805
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.27129018782269604,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3899,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10262542217969894,
      "step": 780,
      "valid_targets_mean": 4038.1,
      "valid_targets_min": 1461
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5130672799557714,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3782,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.40260475873947144,
      "step": 785,
      "valid_targets_mean": 3870.8,
      "valid_targets_min": 1612
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.40260475873947144,
      "step": 785,
      "total_flos": 1.3461651925678162e+18,
      "train_loss": 0.41955339954157544,
      "train_runtime": 17506.3111,
      "train_samples_per_second": 2.856,
      "train_steps_per_second": 0.045,
      "valid_targets_mean": 3870.8,
      "valid_targets_min": 1612
    }
  ],
  "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": 1.3461651925678162e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}