{
  "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": 7.684021027819405,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.9462,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.151728093624115,
      "step": 5,
      "valid_targets_mean": 2826.7,
      "valid_targets_min": 592
    },
    {
      "epoch": 0.064,
      "grad_norm": 4.315659756529689,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.9194,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.25856083631515503,
      "step": 10,
      "valid_targets_mean": 5619.8,
      "valid_targets_min": 496
    },
    {
      "epoch": 0.096,
      "grad_norm": 2.8821577216261813,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.8634,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15587352216243744,
      "step": 15,
      "valid_targets_mean": 2038.3,
      "valid_targets_min": 616
    },
    {
      "epoch": 0.128,
      "grad_norm": 1.2679482794656005,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.8226,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.20857703685760498,
      "step": 20,
      "valid_targets_mean": 3938.2,
      "valid_targets_min": 1116
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.9612116522947592,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.7878,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.24014708399772644,
      "step": 25,
      "valid_targets_mean": 4070.2,
      "valid_targets_min": 693
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.671837378056013,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.7555,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15791478753089905,
      "step": 30,
      "valid_targets_mean": 4183.7,
      "valid_targets_min": 831
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.5382193797381088,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.7207,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.18812517821788788,
      "step": 35,
      "valid_targets_mean": 4437.3,
      "valid_targets_min": 572
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.5061979704759964,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.7071,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15051886439323425,
      "step": 40,
      "valid_targets_mean": 3659.9,
      "valid_targets_min": 691
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.38930620971368396,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.6642,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1958039104938507,
      "step": 45,
      "valid_targets_mean": 4482.8,
      "valid_targets_min": 785
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.40141694725581395,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.6582,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.19386473298072815,
      "step": 50,
      "valid_targets_mean": 4235.5,
      "valid_targets_min": 418
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.4147081064704592,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.644,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14404885470867157,
      "step": 55,
      "valid_targets_mean": 2856.6,
      "valid_targets_min": 660
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.3614259078286061,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.6387,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13555730879306793,
      "step": 60,
      "valid_targets_mean": 3201.2,
      "valid_targets_min": 447
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.3209866041795585,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.6092,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1452401727437973,
      "step": 65,
      "valid_targets_mean": 4834.0,
      "valid_targets_min": 767
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.3544266364187076,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.6023,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1389968991279602,
      "step": 70,
      "valid_targets_mean": 3426.3,
      "valid_targets_min": 492
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.39854699182577796,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.6151,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15443900227546692,
      "step": 75,
      "valid_targets_mean": 3532.5,
      "valid_targets_min": 851
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.35155304524235664,
      "learning_rate": 4e-05,
      "loss": 0.5893,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16085922718048096,
      "step": 80,
      "valid_targets_mean": 4412.2,
      "valid_targets_min": 535
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.36749261955558693,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.5773,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1152658611536026,
      "step": 85,
      "valid_targets_mean": 2273.1,
      "valid_targets_min": 433
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.3265505623639361,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.5777,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14001351594924927,
      "step": 90,
      "valid_targets_mean": 4234.2,
      "valid_targets_min": 578
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.33247581252126013,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.5804,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13451600074768066,
      "step": 95,
      "valid_targets_mean": 4123.6,
      "valid_targets_min": 583
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.31435706508260297,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.5653,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.21264426410198212,
      "step": 100,
      "valid_targets_mean": 7514.6,
      "valid_targets_min": 874
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.46626225277683786,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.5564,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1418621838092804,
      "step": 105,
      "valid_targets_mean": 3887.5,
      "valid_targets_min": 589
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.32985834686353044,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.5564,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13883614540100098,
      "step": 110,
      "valid_targets_mean": 4343.2,
      "valid_targets_min": 724
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.3354634192839252,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.5632,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11766032874584198,
      "step": 115,
      "valid_targets_mean": 3457.9,
      "valid_targets_min": 631
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.3291218980403151,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.5356,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16707633435726166,
      "step": 120,
      "valid_targets_mean": 4794.7,
      "valid_targets_min": 1011
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.3821052903817402,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.5479,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15846571326255798,
      "step": 125,
      "valid_targets_mean": 4849.6,
      "valid_targets_min": 882
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.35955259917565235,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.5292,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12442129105329514,
      "step": 130,
      "valid_targets_mean": 4395.4,
      "valid_targets_min": 617
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.3502103147616533,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.5211,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12329226732254028,
      "step": 135,
      "valid_targets_mean": 3434.2,
      "valid_targets_min": 733
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.3637046459867952,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.5336,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1742524653673172,
      "step": 140,
      "valid_targets_mean": 5235.8,
      "valid_targets_min": 828
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.35235907046102,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.5415,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1258586049079895,
      "step": 145,
      "valid_targets_mean": 3263.0,
      "valid_targets_min": 735
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.32190458967738744,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.5297,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09791816771030426,
      "step": 150,
      "valid_targets_mean": 3830.8,
      "valid_targets_min": 628
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.30079460588774143,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.5439,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10519341379404068,
      "step": 155,
      "valid_targets_mean": 3723.4,
      "valid_targets_min": 588
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.3719413339791788,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.5234,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12874507904052734,
      "step": 160,
      "valid_targets_mean": 5348.6,
      "valid_targets_min": 672
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.3838640800620298,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.5178,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1573619544506073,
      "step": 165,
      "valid_targets_mean": 4277.6,
      "valid_targets_min": 634
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.3139322780346497,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.5287,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16016295552253723,
      "step": 170,
      "valid_targets_mean": 6014.3,
      "valid_targets_min": 739
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.3138254774366607,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.5132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1376572847366333,
      "step": 175,
      "valid_targets_mean": 4934.5,
      "valid_targets_min": 686
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.3932167507717449,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.5393,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16297011077404022,
      "step": 180,
      "valid_targets_mean": 3801.5,
      "valid_targets_min": 546
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.3769579341714615,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.5236,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.18043404817581177,
      "step": 185,
      "valid_targets_mean": 5649.6,
      "valid_targets_min": 647
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.4093275086846018,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.5173,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1410539150238037,
      "step": 190,
      "valid_targets_mean": 4146.4,
      "valid_targets_min": 803
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.36602893155877353,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.5132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0945071429014206,
      "step": 195,
      "valid_targets_mean": 2438.4,
      "valid_targets_min": 570
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.3343967507869978,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.5142,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0918922945857048,
      "step": 200,
      "valid_targets_mean": 4188.4,
      "valid_targets_min": 502
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.3894109753527522,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.5048,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1514074206352234,
      "step": 205,
      "valid_targets_mean": 5301.4,
      "valid_targets_min": 652
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.32859496907018143,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.5115,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10357268899679184,
      "step": 210,
      "valid_targets_mean": 3774.9,
      "valid_targets_min": 958
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.35246638890835796,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.5119,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13515760004520416,
      "step": 215,
      "valid_targets_mean": 4644.8,
      "valid_targets_min": 629
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.33537810996957285,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.5062,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10893318057060242,
      "step": 220,
      "valid_targets_mean": 3986.8,
      "valid_targets_min": 663
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.3712384511413852,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.489,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10876543819904327,
      "step": 225,
      "valid_targets_mean": 3914.3,
      "valid_targets_min": 618
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.31302280882534694,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.4969,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1476151943206787,
      "step": 230,
      "valid_targets_mean": 6176.2,
      "valid_targets_min": 663
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.3554679049086009,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.5072,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09821851551532745,
      "step": 235,
      "valid_targets_mean": 2999.3,
      "valid_targets_min": 858
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.31419523942228644,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.5018,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09154374152421951,
      "step": 240,
      "valid_targets_mean": 3513.5,
      "valid_targets_min": 707
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.3990336294057638,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.5188,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12577299773693085,
      "step": 245,
      "valid_targets_mean": 3209.4,
      "valid_targets_min": 634
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.3845735456163243,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.5057,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1090874969959259,
      "step": 250,
      "valid_targets_mean": 3840.0,
      "valid_targets_min": 807
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.34337043861638583,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.4994,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14603865146636963,
      "step": 255,
      "valid_targets_mean": 5013.7,
      "valid_targets_min": 899
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.3713335418415261,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.5019,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0972067192196846,
      "step": 260,
      "valid_targets_mean": 2316.5,
      "valid_targets_min": 708
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.3796820344471861,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.4965,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1342923641204834,
      "step": 265,
      "valid_targets_mean": 4505.1,
      "valid_targets_min": 715
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.3919217748723875,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.4976,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13651490211486816,
      "step": 270,
      "valid_targets_mean": 4616.6,
      "valid_targets_min": 678
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.3728646075515707,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.4931,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1416204571723938,
      "step": 275,
      "valid_targets_mean": 4793.6,
      "valid_targets_min": 651
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.3504724495840077,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.491,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13293160498142242,
      "step": 280,
      "valid_targets_mean": 5146.3,
      "valid_targets_min": 741
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.3226793223611923,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.5079,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1446695625782013,
      "step": 285,
      "valid_targets_mean": 5345.0,
      "valid_targets_min": 629
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.3574075903283297,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.5034,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11888499557971954,
      "step": 290,
      "valid_targets_mean": 3430.7,
      "valid_targets_min": 654
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.31689085907125186,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.4984,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08475592732429504,
      "step": 295,
      "valid_targets_mean": 3068.2,
      "valid_targets_min": 606
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.32836662379704723,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.4974,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13879406452178955,
      "step": 300,
      "valid_targets_mean": 5841.6,
      "valid_targets_min": 894
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.3372646342505708,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.4968,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.18043114244937897,
      "step": 305,
      "valid_targets_mean": 5560.9,
      "valid_targets_min": 839
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.3649810850111879,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.4971,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.143715500831604,
      "step": 310,
      "valid_targets_mean": 4211.9,
      "valid_targets_min": 717
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.3983193457650017,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.5354,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12572535872459412,
      "step": 315,
      "valid_targets_mean": 2996.9,
      "valid_targets_min": 474
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.36664643318939555,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.4907,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10046903789043427,
      "step": 320,
      "valid_targets_mean": 3342.6,
      "valid_targets_min": 449
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.3702165124978452,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.4854,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15069612860679626,
      "step": 325,
      "valid_targets_mean": 4375.1,
      "valid_targets_min": 970
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.37848234945732695,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.4834,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1191079393029213,
      "step": 330,
      "valid_targets_mean": 3311.8,
      "valid_targets_min": 547
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.3664074498463561,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.4922,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06949648261070251,
      "step": 335,
      "valid_targets_mean": 2625.5,
      "valid_targets_min": 646
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.33635470980471455,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.4855,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12948651611804962,
      "step": 340,
      "valid_targets_mean": 5193.2,
      "valid_targets_min": 785
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.3772883851831424,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.474,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14694371819496155,
      "step": 345,
      "valid_targets_mean": 4448.8,
      "valid_targets_min": 713
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.3644965519784912,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.4745,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16046449542045593,
      "step": 350,
      "valid_targets_mean": 5504.3,
      "valid_targets_min": 479
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.35661151067402347,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.4745,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11481410264968872,
      "step": 355,
      "valid_targets_mean": 3303.3,
      "valid_targets_min": 579
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.3443551079928496,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.4803,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09265729784965515,
      "step": 360,
      "valid_targets_mean": 3211.2,
      "valid_targets_min": 701
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.33024838333473727,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.4827,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09647121280431747,
      "step": 365,
      "valid_targets_mean": 3607.2,
      "valid_targets_min": 776
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.3696473132279038,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.471,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14742985367774963,
      "step": 370,
      "valid_targets_mean": 5429.0,
      "valid_targets_min": 698
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.3562973430615155,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.4827,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.089414581656456,
      "step": 375,
      "valid_targets_mean": 2608.9,
      "valid_targets_min": 608
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.3407650809642004,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.476,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11797565221786499,
      "step": 380,
      "valid_targets_mean": 4561.6,
      "valid_targets_min": 883
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.333562991638242,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.47,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14663121104240417,
      "step": 385,
      "valid_targets_mean": 6017.0,
      "valid_targets_min": 559
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.3168575731266746,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.4764,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13627192378044128,
      "step": 390,
      "valid_targets_mean": 5136.8,
      "valid_targets_min": 575
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.3783190721732701,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.4808,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13549643754959106,
      "step": 395,
      "valid_targets_mean": 3961.8,
      "valid_targets_min": 654
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.3985733386715135,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.4891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07634571194648743,
      "step": 400,
      "valid_targets_mean": 2903.4,
      "valid_targets_min": 561
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.34530197059474554,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.4756,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08810963481664658,
      "step": 405,
      "valid_targets_mean": 2997.7,
      "valid_targets_min": 639
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.28530664993093097,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.4659,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11760587990283966,
      "step": 410,
      "valid_targets_mean": 5408.6,
      "valid_targets_min": 618
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.30385632137630963,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.482,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13619443774223328,
      "step": 415,
      "valid_targets_mean": 5362.3,
      "valid_targets_min": 603
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.35491269233908346,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.4743,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09804345667362213,
      "step": 420,
      "valid_targets_mean": 2615.6,
      "valid_targets_min": 644
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.35701706432054836,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.4838,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14509597420692444,
      "step": 425,
      "valid_targets_mean": 4059.8,
      "valid_targets_min": 815
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.29447082854824025,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.4696,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13219314813613892,
      "step": 430,
      "valid_targets_mean": 6867.2,
      "valid_targets_min": 613
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.3405190866493231,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.479,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13819676637649536,
      "step": 435,
      "valid_targets_mean": 5100.7,
      "valid_targets_min": 723
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.3192777853267772,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.4785,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11653417348861694,
      "step": 440,
      "valid_targets_mean": 3380.3,
      "valid_targets_min": 813
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.37687082035578107,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.4778,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09297601133584976,
      "step": 445,
      "valid_targets_mean": 3103.0,
      "valid_targets_min": 608
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.30763919174062704,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.4792,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10154986381530762,
      "step": 450,
      "valid_targets_mean": 4853.2,
      "valid_targets_min": 878
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.31724492152476136,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.4728,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11423113942146301,
      "step": 455,
      "valid_targets_mean": 4161.6,
      "valid_targets_min": 744
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.31997067824732267,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.4778,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11144183576107025,
      "step": 460,
      "valid_targets_mean": 4194.6,
      "valid_targets_min": 499
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.3538431941795157,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.4743,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13119390606880188,
      "step": 465,
      "valid_targets_mean": 3995.3,
      "valid_targets_min": 773
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.33600503224781925,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.4768,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09177535027265549,
      "step": 470,
      "valid_targets_mean": 3161.3,
      "valid_targets_min": 540
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.3493519097385825,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.4723,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12285567820072174,
      "step": 475,
      "valid_targets_mean": 3352.7,
      "valid_targets_min": 625
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.31276946697938807,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.4708,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10380571335554123,
      "step": 480,
      "valid_targets_mean": 4585.8,
      "valid_targets_min": 538
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.29872261029710623,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.4566,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12298127263784409,
      "step": 485,
      "valid_targets_mean": 6115.2,
      "valid_targets_min": 727
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.34689061804276644,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.4636,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12354570627212524,
      "step": 490,
      "valid_targets_mean": 4252.8,
      "valid_targets_min": 778
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.38032979180019366,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.4657,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12418144196271896,
      "step": 495,
      "valid_targets_mean": 3636.5,
      "valid_targets_min": 496
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.3431853386357883,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.4715,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09321945160627365,
      "step": 500,
      "valid_targets_mean": 2691.8,
      "valid_targets_min": 675
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.30947818564912566,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.4707,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16295385360717773,
      "step": 505,
      "valid_targets_mean": 6233.9,
      "valid_targets_min": 564
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.338142930798456,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.4713,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0744905024766922,
      "step": 510,
      "valid_targets_mean": 2177.6,
      "valid_targets_min": 565
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.3404807432073367,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.4525,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1156606525182724,
      "step": 515,
      "valid_targets_mean": 4082.6,
      "valid_targets_min": 306
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.40982359175260846,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.4691,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14627587795257568,
      "step": 520,
      "valid_targets_mean": 3950.3,
      "valid_targets_min": 778
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.36041221063560147,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.4624,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10033407807350159,
      "step": 525,
      "valid_targets_mean": 3262.5,
      "valid_targets_min": 971
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.3214728011282713,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.4602,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11894414573907852,
      "step": 530,
      "valid_targets_mean": 5553.6,
      "valid_targets_min": 973
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.3386089697033676,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.4506,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10617012530565262,
      "step": 535,
      "valid_targets_mean": 3929.0,
      "valid_targets_min": 763
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.33946770153693895,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.4621,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08171704411506653,
      "step": 540,
      "valid_targets_mean": 2863.6,
      "valid_targets_min": 328
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.2976297864581848,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.46,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1289682239294052,
      "step": 545,
      "valid_targets_mean": 4881.0,
      "valid_targets_min": 886
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.312424429008915,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.4672,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13805308938026428,
      "step": 550,
      "valid_targets_mean": 5005.2,
      "valid_targets_min": 715
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.2999713019596007,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.4606,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08807637542486191,
      "step": 555,
      "valid_targets_mean": 3781.8,
      "valid_targets_min": 608
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.344815000465373,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.4624,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08266343176364899,
      "step": 560,
      "valid_targets_mean": 3264.2,
      "valid_targets_min": 573
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.33309014723392927,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.4753,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08952188491821289,
      "step": 565,
      "valid_targets_mean": 3132.6,
      "valid_targets_min": 709
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.31695965289849937,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.4576,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09465669840574265,
      "step": 570,
      "valid_targets_mean": 3652.5,
      "valid_targets_min": 495
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.3490161980208642,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.4574,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12921452522277832,
      "step": 575,
      "valid_targets_mean": 4193.8,
      "valid_targets_min": 985
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.35392398907941525,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.4696,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12406637519598007,
      "step": 580,
      "valid_targets_mean": 4371.9,
      "valid_targets_min": 601
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.3887032347522501,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.4547,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11641094833612442,
      "step": 585,
      "valid_targets_mean": 3000.5,
      "valid_targets_min": 351
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.34317600174415436,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.4658,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1424415111541748,
      "step": 590,
      "valid_targets_mean": 4481.1,
      "valid_targets_min": 898
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.3237463389809962,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.4675,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11732232570648193,
      "step": 595,
      "valid_targets_mean": 4616.7,
      "valid_targets_min": 749
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.33236148911135943,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.4523,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12217555940151215,
      "step": 600,
      "valid_targets_mean": 4250.4,
      "valid_targets_min": 713
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.3414166296388241,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.4671,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1075812429189682,
      "step": 605,
      "valid_targets_mean": 4430.2,
      "valid_targets_min": 817
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.3195891738556361,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.4557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12964078783988953,
      "step": 610,
      "valid_targets_mean": 4666.1,
      "valid_targets_min": 670
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.30545122838124106,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.4618,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15205875039100647,
      "step": 615,
      "valid_targets_mean": 6188.0,
      "valid_targets_min": 765
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.347216962814736,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.4698,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07474929094314575,
      "step": 620,
      "valid_targets_mean": 2098.2,
      "valid_targets_min": 559
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.28211289565938785,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.4606,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1354532539844513,
      "step": 625,
      "valid_targets_mean": 6751.6,
      "valid_targets_min": 598
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.32132201707056995,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.4674,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1583113968372345,
      "step": 630,
      "valid_targets_mean": 5615.2,
      "valid_targets_min": 679
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.36186093521738927,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.4576,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10461312532424927,
      "step": 635,
      "valid_targets_mean": 3433.2,
      "valid_targets_min": 536
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.31714500982976623,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.4599,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12880629301071167,
      "step": 640,
      "valid_targets_mean": 5376.7,
      "valid_targets_min": 614
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.3310278200604046,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.4622,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10139229893684387,
      "step": 645,
      "valid_targets_mean": 2878.0,
      "valid_targets_min": 661
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.3405506801031202,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.4596,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13550347089767456,
      "step": 650,
      "valid_targets_mean": 4592.4,
      "valid_targets_min": 746
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.31232924096862863,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.4521,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15010719001293182,
      "step": 655,
      "valid_targets_mean": 6127.2,
      "valid_targets_min": 934
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.32990716958733296,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.4612,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12817400693893433,
      "step": 660,
      "valid_targets_mean": 5087.4,
      "valid_targets_min": 931
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.3594633114396284,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.4449,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07827775180339813,
      "step": 665,
      "valid_targets_mean": 2190.4,
      "valid_targets_min": 758
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.3284356121316824,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.4423,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09074963629245758,
      "step": 670,
      "valid_targets_mean": 3644.5,
      "valid_targets_min": 578
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.32901414159059233,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.4452,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1113700270652771,
      "step": 675,
      "valid_targets_mean": 3857.1,
      "valid_targets_min": 678
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.31453317906103917,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.4583,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10023031383752823,
      "step": 680,
      "valid_targets_mean": 4228.3,
      "valid_targets_min": 696
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.28948536548009773,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.4497,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11549019068479538,
      "step": 685,
      "valid_targets_mean": 5232.1,
      "valid_targets_min": 798
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.3068531051896559,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.4587,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11238421499729156,
      "step": 690,
      "valid_targets_mean": 4504.8,
      "valid_targets_min": 713
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.34758225609480875,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.4639,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11710608005523682,
      "step": 695,
      "valid_targets_mean": 3600.1,
      "valid_targets_min": 780
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.3500273648227193,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.4593,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1051868200302124,
      "step": 700,
      "valid_targets_mean": 3240.8,
      "valid_targets_min": 670
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.403657898277183,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.4692,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1348893642425537,
      "step": 705,
      "valid_targets_mean": 4384.1,
      "valid_targets_min": 583
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.33854447664044307,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.4598,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11955916881561279,
      "step": 710,
      "valid_targets_mean": 3573.9,
      "valid_targets_min": 676
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.33198299669845865,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.4712,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12793362140655518,
      "step": 715,
      "valid_targets_mean": 4069.8,
      "valid_targets_min": 516
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.33749675621254677,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.4713,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10581079125404358,
      "step": 720,
      "valid_targets_mean": 4194.9,
      "valid_targets_min": 827
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.32552916542542465,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.4499,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13471418619155884,
      "step": 725,
      "valid_targets_mean": 4708.9,
      "valid_targets_min": 675
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.3407528296199396,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.4669,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10393006354570389,
      "step": 730,
      "valid_targets_mean": 4275.6,
      "valid_targets_min": 593
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.34056838452021004,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.4598,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15185418725013733,
      "step": 735,
      "valid_targets_mean": 4414.4,
      "valid_targets_min": 875
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.31855375982058554,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.4553,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1116475909948349,
      "step": 740,
      "valid_targets_mean": 4347.5,
      "valid_targets_min": 573
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.34631116302634884,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.4513,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09804511815309525,
      "step": 745,
      "valid_targets_mean": 3213.0,
      "valid_targets_min": 1013
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.30322542159004084,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.4607,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10131211578845978,
      "step": 750,
      "valid_targets_mean": 3960.2,
      "valid_targets_min": 426
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.3552281533109692,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.4454,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10585194826126099,
      "step": 755,
      "valid_targets_mean": 5761.2,
      "valid_targets_min": 624
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.33098999146421887,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.4532,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1376958042383194,
      "step": 760,
      "valid_targets_mean": 4733.8,
      "valid_targets_min": 754
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.3404239677960933,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.4599,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10137306153774261,
      "step": 765,
      "valid_targets_mean": 3575.0,
      "valid_targets_min": 418
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.2689340792581925,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.4529,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1110411062836647,
      "step": 770,
      "valid_targets_mean": 5378.4,
      "valid_targets_min": 812
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.3467018787004929,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.4713,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07618623971939087,
      "step": 775,
      "valid_targets_mean": 2050.8,
      "valid_targets_min": 583
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.2975987200384198,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.4476,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09185172617435455,
      "step": 780,
      "valid_targets_mean": 3924.0,
      "valid_targets_min": 869
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.7179511244062344,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.4444,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.42960670590400696,
      "step": 785,
      "valid_targets_mean": 3849.6,
      "valid_targets_min": 730
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.42960670590400696,
      "step": 785,
      "total_flos": 1.3335511329281147e+18,
      "train_loss": 0.5088972483470942,
      "train_runtime": 21724.3827,
      "train_samples_per_second": 2.299,
      "train_steps_per_second": 0.036,
      "valid_targets_mean": 3849.6,
      "valid_targets_min": 730
    }
  ],
  "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": 1.3335511329281147e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}