{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 785,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032,
      "grad_norm": 4.8957957096867695,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.6553,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15291616320610046,
      "step": 5,
      "valid_targets_mean": 5641.3,
      "valid_targets_min": 1301
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.5543027927662174,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.6826,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13828763365745544,
      "step": 10,
      "valid_targets_mean": 5444.6,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.096,
      "grad_norm": 1.898175396562219,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.6055,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16061976552009583,
      "step": 15,
      "valid_targets_mean": 5732.6,
      "valid_targets_min": 2447
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.7687661316622517,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.5813,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13461264967918396,
      "step": 20,
      "valid_targets_mean": 5091.4,
      "valid_targets_min": 834
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6650915724696022,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.5519,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10425147414207458,
      "step": 25,
      "valid_targets_mean": 3950.8,
      "valid_targets_min": 627
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.4513449568541331,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.5304,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11923600733280182,
      "step": 30,
      "valid_targets_mean": 5288.8,
      "valid_targets_min": 695
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.3931346625843204,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5169,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1516682505607605,
      "step": 35,
      "valid_targets_mean": 5724.2,
      "valid_targets_min": 1349
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.29442462619191745,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.4813,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09292076528072357,
      "step": 40,
      "valid_targets_mean": 5563.2,
      "valid_targets_min": 1156
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.27304562511080793,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.4649,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08254602551460266,
      "step": 45,
      "valid_targets_mean": 4044.0,
      "valid_targets_min": 505
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.24150465933338938,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.4587,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08485572040081024,
      "step": 50,
      "valid_targets_mean": 4775.4,
      "valid_targets_min": 891
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.23104735718228317,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.4545,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11856499314308167,
      "step": 55,
      "valid_targets_mean": 6933.8,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.2248853945555057,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.4294,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10473228991031647,
      "step": 60,
      "valid_targets_mean": 5542.1,
      "valid_targets_min": 748
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.2642286257956458,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.4524,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09943065047264099,
      "step": 65,
      "valid_targets_mean": 4435.2,
      "valid_targets_min": 666
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.24344645114272737,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.4223,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08963267505168915,
      "step": 70,
      "valid_targets_mean": 5001.8,
      "valid_targets_min": 770
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2151719743915404,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.4355,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09131129086017609,
      "step": 75,
      "valid_targets_mean": 5227.5,
      "valid_targets_min": 442
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.2583130508350391,
      "learning_rate": 4e-05,
      "loss": 0.4259,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09952086955308914,
      "step": 80,
      "valid_targets_mean": 4703.3,
      "valid_targets_min": 712
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.22535297691684805,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.409,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09645680338144302,
      "step": 85,
      "valid_targets_mean": 6303.9,
      "valid_targets_min": 659
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.2400640865792244,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.4086,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10337771475315094,
      "step": 90,
      "valid_targets_mean": 4726.8,
      "valid_targets_min": 670
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.2197260241311898,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4113,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10253070294857025,
      "step": 95,
      "valid_targets_mean": 5971.6,
      "valid_targets_min": 974
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.25215977200006373,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.4031,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10277324914932251,
      "step": 100,
      "valid_targets_mean": 5096.6,
      "valid_targets_min": 886
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.23520983500395826,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.4058,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12166647613048553,
      "step": 105,
      "valid_targets_mean": 6824.0,
      "valid_targets_min": 760
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.35704329215585356,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4008,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11661948263645172,
      "step": 110,
      "valid_targets_mean": 6983.6,
      "valid_targets_min": 1467
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.23288396565696193,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.4002,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11187885701656342,
      "step": 115,
      "valid_targets_mean": 5738.2,
      "valid_targets_min": 576
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.21916732010829937,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.3991,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10342550277709961,
      "step": 120,
      "valid_targets_mean": 5972.9,
      "valid_targets_min": 1069
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.23423531421676796,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4014,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0816822424530983,
      "step": 125,
      "valid_targets_mean": 6150.6,
      "valid_targets_min": 1019
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.22943283836884876,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.3996,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10121267288923264,
      "step": 130,
      "valid_targets_mean": 5070.9,
      "valid_targets_min": 687
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.2226691048634411,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.405,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0976373553276062,
      "step": 135,
      "valid_targets_mean": 4816.9,
      "valid_targets_min": 496
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.23818174232330994,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.3891,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08187875896692276,
      "step": 140,
      "valid_targets_mean": 5070.4,
      "valid_targets_min": 729
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.216733332557618,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.3737,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09527260065078735,
      "step": 145,
      "valid_targets_mean": 6045.0,
      "valid_targets_min": 812
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.25910736011188307,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.3862,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11887587606906891,
      "step": 150,
      "valid_targets_mean": 5690.5,
      "valid_targets_min": 987
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.22890462810601953,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.3812,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09058446437120438,
      "step": 155,
      "valid_targets_mean": 5883.9,
      "valid_targets_min": 676
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.2399097142566408,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.3923,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08804625272750854,
      "step": 160,
      "valid_targets_mean": 4297.6,
      "valid_targets_min": 575
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.22274200226906252,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.378,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09087201207876205,
      "step": 165,
      "valid_targets_mean": 5597.7,
      "valid_targets_min": 683
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.22463331974536,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.385,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07766157388687134,
      "step": 170,
      "valid_targets_mean": 4806.2,
      "valid_targets_min": 858
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.23183069838495832,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.375,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09167921543121338,
      "step": 175,
      "valid_targets_mean": 4686.6,
      "valid_targets_min": 535
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.22489150071342387,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.3772,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1004074215888977,
      "step": 180,
      "valid_targets_mean": 5953.7,
      "valid_targets_min": 1237
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.2514549482978997,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.3787,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0844147652387619,
      "step": 185,
      "valid_targets_mean": 4131.6,
      "valid_targets_min": 826
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.22207122707715768,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.3804,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09314845502376556,
      "step": 190,
      "valid_targets_mean": 5937.2,
      "valid_targets_min": 735
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.2259033766041621,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.3689,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10375134646892548,
      "step": 195,
      "valid_targets_mean": 6962.8,
      "valid_targets_min": 2016
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.2305419244650642,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.3786,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09479468315839767,
      "step": 200,
      "valid_targets_mean": 5437.2,
      "valid_targets_min": 824
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.27735045091104826,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.3797,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07483960688114166,
      "step": 205,
      "valid_targets_mean": 5013.6,
      "valid_targets_min": 712
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.2385057691021884,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.3838,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1109888106584549,
      "step": 210,
      "valid_targets_mean": 6360.2,
      "valid_targets_min": 1310
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.2420982570417425,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.3767,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07871495187282562,
      "step": 215,
      "valid_targets_mean": 4264.9,
      "valid_targets_min": 544
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.23672388185810728,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.3771,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10654886066913605,
      "step": 220,
      "valid_targets_mean": 6411.9,
      "valid_targets_min": 615
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.24173970680985113,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.3718,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10277239978313446,
      "step": 225,
      "valid_targets_mean": 5812.1,
      "valid_targets_min": 747
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.2033364238458767,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.3761,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09572498500347137,
      "step": 230,
      "valid_targets_mean": 6630.9,
      "valid_targets_min": 1951
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.25927123982700884,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.3743,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1029035747051239,
      "step": 235,
      "valid_targets_mean": 5665.6,
      "valid_targets_min": 946
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.22003278861461334,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.3752,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0721648633480072,
      "step": 240,
      "valid_targets_mean": 4442.8,
      "valid_targets_min": 627
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.22700896794253264,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.3735,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09452672302722931,
      "step": 245,
      "valid_targets_mean": 6673.9,
      "valid_targets_min": 1380
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.2644829252243767,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.3951,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.095633365213871,
      "step": 250,
      "valid_targets_mean": 6358.1,
      "valid_targets_min": 988
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.22094093436296233,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.3742,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07428319752216339,
      "step": 255,
      "valid_targets_mean": 5924.9,
      "valid_targets_min": 699
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.23457949026801236,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.3701,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09160979837179184,
      "step": 260,
      "valid_targets_mean": 6274.9,
      "valid_targets_min": 840
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.22631399215346612,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.3697,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08914121985435486,
      "step": 265,
      "valid_targets_mean": 4970.8,
      "valid_targets_min": 717
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.2281946190221415,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.3628,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10558171570301056,
      "step": 270,
      "valid_targets_mean": 6200.3,
      "valid_targets_min": 834
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.22934517335485496,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.3596,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09920540452003479,
      "step": 275,
      "valid_targets_mean": 5179.2,
      "valid_targets_min": 1147
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.22979248966153246,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.3708,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10385757684707642,
      "step": 280,
      "valid_targets_mean": 6714.1,
      "valid_targets_min": 1588
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.23343247792090419,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.3661,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09786373376846313,
      "step": 285,
      "valid_targets_mean": 5761.8,
      "valid_targets_min": 994
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.21309462758952905,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.3703,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08539576828479767,
      "step": 290,
      "valid_targets_mean": 6083.4,
      "valid_targets_min": 672
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.22436695662964995,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.3747,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0956336110830307,
      "step": 295,
      "valid_targets_mean": 5751.5,
      "valid_targets_min": 805
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.21701551890786766,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.3671,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10707379877567291,
      "step": 300,
      "valid_targets_mean": 6680.9,
      "valid_targets_min": 718
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.21694076717005004,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.3659,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08210191875696182,
      "step": 305,
      "valid_targets_mean": 5516.9,
      "valid_targets_min": 605
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.2510602866355029,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.3768,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08098435401916504,
      "step": 310,
      "valid_targets_mean": 4204.3,
      "valid_targets_min": 663
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.22702991129592767,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.3731,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09287005662918091,
      "step": 315,
      "valid_targets_mean": 6363.2,
      "valid_targets_min": 1309
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.22469051767149656,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.3709,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1076088547706604,
      "step": 320,
      "valid_targets_mean": 6793.4,
      "valid_targets_min": 842
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.248035025740067,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.3627,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10270263254642487,
      "step": 325,
      "valid_targets_mean": 7012.6,
      "valid_targets_min": 962
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.21743003349061876,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.3534,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07702721655368805,
      "step": 330,
      "valid_targets_mean": 6053.9,
      "valid_targets_min": 631
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.2295885203480888,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.3569,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07063645124435425,
      "step": 335,
      "valid_targets_mean": 4342.3,
      "valid_targets_min": 831
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.23084840199181872,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.3583,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11598502099514008,
      "step": 340,
      "valid_targets_mean": 6150.0,
      "valid_targets_min": 798
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.21765958833700927,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.3579,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07977087050676346,
      "step": 345,
      "valid_targets_mean": 6211.4,
      "valid_targets_min": 1315
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.2285077765285173,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.3585,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10263460129499435,
      "step": 350,
      "valid_targets_mean": 6624.2,
      "valid_targets_min": 773
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.23339941948818674,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.3577,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07275256514549255,
      "step": 355,
      "valid_targets_mean": 4593.9,
      "valid_targets_min": 779
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.2502105294796161,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.3559,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08883202075958252,
      "step": 360,
      "valid_targets_mean": 5197.9,
      "valid_targets_min": 720
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.21904117387811056,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.3647,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08322088420391083,
      "step": 365,
      "valid_targets_mean": 5329.1,
      "valid_targets_min": 1164
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.2109000942177344,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.3627,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09816489368677139,
      "step": 370,
      "valid_targets_mean": 6357.5,
      "valid_targets_min": 843
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.21088724996195515,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3604,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09968498349189758,
      "step": 375,
      "valid_targets_mean": 6617.5,
      "valid_targets_min": 475
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.22879218090156908,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.3596,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07352171838283539,
      "step": 380,
      "valid_targets_mean": 4741.2,
      "valid_targets_min": 724
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.23819106263312376,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.3576,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11139841377735138,
      "step": 385,
      "valid_targets_mean": 6512.4,
      "valid_targets_min": 1490
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.23375791982174612,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.3562,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08265475928783417,
      "step": 390,
      "valid_targets_mean": 5092.9,
      "valid_targets_min": 603
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.22207111267589996,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.3615,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09710904955863953,
      "step": 395,
      "valid_targets_mean": 6565.7,
      "valid_targets_min": 1146
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.20874730912919356,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.3581,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06831514835357666,
      "step": 400,
      "valid_targets_mean": 4616.6,
      "valid_targets_min": 484
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.23666610436656774,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.364,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08490247279405594,
      "step": 405,
      "valid_targets_mean": 5329.2,
      "valid_targets_min": 589
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.22395567453084383,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.362,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07954832911491394,
      "step": 410,
      "valid_targets_mean": 4936.8,
      "valid_targets_min": 861
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.22538655394597762,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.3605,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08222854137420654,
      "step": 415,
      "valid_targets_mean": 4606.9,
      "valid_targets_min": 615
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.21787726181093003,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.3638,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08987221121788025,
      "step": 420,
      "valid_targets_mean": 5660.1,
      "valid_targets_min": 644
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.23492850405635946,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.3634,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09882394969463348,
      "step": 425,
      "valid_targets_mean": 7214.3,
      "valid_targets_min": 920
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.2549023643633448,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.3628,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10947997123003006,
      "step": 430,
      "valid_targets_mean": 5776.4,
      "valid_targets_min": 585
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.22745799161726885,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.3493,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0875130444765091,
      "step": 435,
      "valid_targets_mean": 5458.2,
      "valid_targets_min": 619
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.19982763676760518,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3514,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09613966941833496,
      "step": 440,
      "valid_targets_mean": 7524.6,
      "valid_targets_min": 997
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.20904147454846914,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.362,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06784340739250183,
      "step": 445,
      "valid_targets_mean": 4123.4,
      "valid_targets_min": 972
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.2272779387271342,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3575,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07681050896644592,
      "step": 450,
      "valid_targets_mean": 4773.1,
      "valid_targets_min": 535
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.22422569191376415,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.3462,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09543701261281967,
      "step": 455,
      "valid_targets_mean": 4987.8,
      "valid_targets_min": 956
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.2127300411651044,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.3543,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09453082084655762,
      "step": 460,
      "valid_targets_mean": 5574.0,
      "valid_targets_min": 494
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.21317603801912732,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.3513,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0802430734038353,
      "step": 465,
      "valid_targets_mean": 6034.9,
      "valid_targets_min": 783
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.2255925029458408,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.3622,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09638278931379318,
      "step": 470,
      "valid_targets_mean": 5085.6,
      "valid_targets_min": 924
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.20294192198907773,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.3577,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10205837339162827,
      "step": 475,
      "valid_targets_mean": 7576.9,
      "valid_targets_min": 2449
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.222788326913705,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.3595,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09679550677537918,
      "step": 480,
      "valid_targets_mean": 6337.4,
      "valid_targets_min": 597
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.21150667897316444,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.3448,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08647511899471283,
      "step": 485,
      "valid_targets_mean": 6353.4,
      "valid_targets_min": 839
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.21495447593089753,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.359,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10146178305149078,
      "step": 490,
      "valid_targets_mean": 6548.2,
      "valid_targets_min": 560
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.2211910578834333,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.3575,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08298438042402267,
      "step": 495,
      "valid_targets_mean": 5273.4,
      "valid_targets_min": 926
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.2157730844708698,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.3557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08530550450086594,
      "step": 500,
      "valid_targets_mean": 5513.8,
      "valid_targets_min": 761
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.20375621347852646,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3531,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09325112402439117,
      "step": 505,
      "valid_targets_mean": 6665.9,
      "valid_targets_min": 610
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.2191739435288772,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3507,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08988802134990692,
      "step": 510,
      "valid_targets_mean": 6045.4,
      "valid_targets_min": 580
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.22813856566463242,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.3527,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09069983661174774,
      "step": 515,
      "valid_targets_mean": 5085.8,
      "valid_targets_min": 724
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.2034647518828202,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3583,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10293670743703842,
      "step": 520,
      "valid_targets_mean": 7327.9,
      "valid_targets_min": 1160
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.25511797039062667,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.3409,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08311108499765396,
      "step": 525,
      "valid_targets_mean": 4695.1,
      "valid_targets_min": 617
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.2062022019289714,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3428,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08815449476242065,
      "step": 530,
      "valid_targets_mean": 6441.4,
      "valid_targets_min": 607
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.217934462967701,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.3518,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09073206782341003,
      "step": 535,
      "valid_targets_mean": 5660.6,
      "valid_targets_min": 760
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.1948122888742335,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.353,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08459735661745071,
      "step": 540,
      "valid_targets_mean": 7381.5,
      "valid_targets_min": 855
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.2060312527089098,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.3424,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08104632049798965,
      "step": 545,
      "valid_targets_mean": 4708.5,
      "valid_targets_min": 631
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.37629522520722347,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.3522,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0980551689863205,
      "step": 550,
      "valid_targets_mean": 5957.8,
      "valid_targets_min": 1062
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.2138782714048621,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3516,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08053107559680939,
      "step": 555,
      "valid_targets_mean": 5044.3,
      "valid_targets_min": 787
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.2183596697848869,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.3583,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10445421189069748,
      "step": 560,
      "valid_targets_mean": 6474.5,
      "valid_targets_min": 544
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.22696842281330923,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.347,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0842842236161232,
      "step": 565,
      "valid_targets_mean": 5337.5,
      "valid_targets_min": 762
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.21309383682105515,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3505,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09089038521051407,
      "step": 570,
      "valid_targets_mean": 5707.0,
      "valid_targets_min": 627
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.2398068187730169,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.3502,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11133481562137604,
      "step": 575,
      "valid_targets_mean": 6219.8,
      "valid_targets_min": 1551
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.22090470996265946,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.3547,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07591253519058228,
      "step": 580,
      "valid_targets_mean": 5110.4,
      "valid_targets_min": 965
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.2109746905963528,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.3505,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09519827365875244,
      "step": 585,
      "valid_targets_mean": 5831.9,
      "valid_targets_min": 748
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.1992738529635182,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3367,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0837312787771225,
      "step": 590,
      "valid_targets_mean": 6252.3,
      "valid_targets_min": 949
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.19876496258222104,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.3524,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07738886028528214,
      "step": 595,
      "valid_targets_mean": 5667.1,
      "valid_targets_min": 1145
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.2093870951378708,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.3526,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06702277064323425,
      "step": 600,
      "valid_targets_mean": 4404.4,
      "valid_targets_min": 673
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.22491046386663305,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3475,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07729315757751465,
      "step": 605,
      "valid_targets_mean": 4666.8,
      "valid_targets_min": 672
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.20573403549765695,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.3626,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09660472720861435,
      "step": 610,
      "valid_targets_mean": 6227.8,
      "valid_targets_min": 633
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.21513206737959598,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.3466,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08919400721788406,
      "step": 615,
      "valid_targets_mean": 6852.4,
      "valid_targets_min": 841
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.19987136693291732,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.3469,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09743735939264297,
      "step": 620,
      "valid_targets_mean": 6217.0,
      "valid_targets_min": 1621
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.22123971957324814,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.3435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09402965009212494,
      "step": 625,
      "valid_targets_mean": 6170.4,
      "valid_targets_min": 996
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.23453973885955678,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3414,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0874083861708641,
      "step": 630,
      "valid_targets_mean": 5678.9,
      "valid_targets_min": 1098
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.21377261233751094,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.3485,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09617868065834045,
      "step": 635,
      "valid_targets_mean": 5427.3,
      "valid_targets_min": 610
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.21522129527154207,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.3497,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0974159762263298,
      "step": 640,
      "valid_targets_mean": 6232.0,
      "valid_targets_min": 1075
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.20617290059221755,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.3437,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08632023632526398,
      "step": 645,
      "valid_targets_mean": 5968.9,
      "valid_targets_min": 679
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.20536855612532723,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.3408,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06517566740512848,
      "step": 650,
      "valid_targets_mean": 4388.8,
      "valid_targets_min": 524
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.22624136593320998,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3465,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10018053650856018,
      "step": 655,
      "valid_targets_mean": 6456.8,
      "valid_targets_min": 1832
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.22977331762904685,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.3478,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07814135402441025,
      "step": 660,
      "valid_targets_mean": 4665.2,
      "valid_targets_min": 900
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.22014586294283828,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.3502,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08816957473754883,
      "step": 665,
      "valid_targets_mean": 5389.8,
      "valid_targets_min": 689
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.21084257754576288,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3429,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07026119530200958,
      "step": 670,
      "valid_targets_mean": 4378.2,
      "valid_targets_min": 633
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.21035491901153253,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3331,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10015265643596649,
      "step": 675,
      "valid_targets_mean": 6495.4,
      "valid_targets_min": 814
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.19453489476838928,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3477,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08647267520427704,
      "step": 680,
      "valid_targets_mean": 6659.5,
      "valid_targets_min": 667
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.21492697505806427,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.3479,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07926177978515625,
      "step": 685,
      "valid_targets_mean": 4461.8,
      "valid_targets_min": 810
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.2075255327131279,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3498,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09420672059059143,
      "step": 690,
      "valid_targets_mean": 6019.9,
      "valid_targets_min": 562
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.21600916108730278,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.3498,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10290233790874481,
      "step": 695,
      "valid_targets_mean": 5586.2,
      "valid_targets_min": 697
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.21249101681367025,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.352,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07698194682598114,
      "step": 700,
      "valid_targets_mean": 4835.4,
      "valid_targets_min": 693
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.2006329639594424,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3417,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09292782098054886,
      "step": 705,
      "valid_targets_mean": 6489.8,
      "valid_targets_min": 989
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.2025250030074837,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.3499,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10066059976816177,
      "step": 710,
      "valid_targets_mean": 6283.6,
      "valid_targets_min": 832
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.2136720770361283,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.3534,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10379557311534882,
      "step": 715,
      "valid_targets_mean": 6109.6,
      "valid_targets_min": 960
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.19061767495330614,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3429,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07969370484352112,
      "step": 720,
      "valid_targets_mean": 5774.0,
      "valid_targets_min": 1182
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.2102063509622617,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.3541,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09941019117832184,
      "step": 725,
      "valid_targets_mean": 6061.4,
      "valid_targets_min": 1024
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.2708209939364983,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.3417,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07649870961904526,
      "step": 730,
      "valid_targets_mean": 5401.0,
      "valid_targets_min": 627
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.20654731085294797,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.3441,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09117791056632996,
      "step": 735,
      "valid_targets_mean": 5738.6,
      "valid_targets_min": 1019
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.206448571776912,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3447,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08553271740674973,
      "step": 740,
      "valid_targets_mean": 5514.2,
      "valid_targets_min": 719
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.18979763950557874,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.3519,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08793099224567413,
      "step": 745,
      "valid_targets_mean": 6579.6,
      "valid_targets_min": 1813
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.19071885468715494,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.3455,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08575663715600967,
      "step": 750,
      "valid_targets_mean": 5970.8,
      "valid_targets_min": 747
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.19940122645940153,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.3435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08477139472961426,
      "step": 755,
      "valid_targets_mean": 6556.9,
      "valid_targets_min": 580
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.2163105931804559,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3484,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07366898655891418,
      "step": 760,
      "valid_targets_mean": 4413.4,
      "valid_targets_min": 762
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.19291624347459255,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3562,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09870646148920059,
      "step": 765,
      "valid_targets_mean": 6162.2,
      "valid_targets_min": 1315
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.22587725365283015,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.3451,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09872078150510788,
      "step": 770,
      "valid_targets_mean": 5996.9,
      "valid_targets_min": 668
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.1996370427448213,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.3453,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09999275952577591,
      "step": 775,
      "valid_targets_mean": 6430.5,
      "valid_targets_min": 1007
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.2114447831817363,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3603,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08034367114305496,
      "step": 780,
      "valid_targets_mean": 4864.9,
      "valid_targets_min": 580
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.416998530647348,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3456,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.3421081006526947,
      "step": 785,
      "valid_targets_mean": 5900.6,
      "valid_targets_min": 645
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.3421081006526947,
      "step": 785,
      "total_flos": 1.879814441009152e+18,
      "train_loss": 0.37729120831580676,
      "train_runtime": 24109.2913,
      "train_samples_per_second": 2.073,
      "train_steps_per_second": 0.033,
      "valid_targets_mean": 5900.6,
      "valid_targets_min": 645
    }
  ],
  "logging_steps": 5,
  "max_steps": 785,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.879814441009152e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}