{
  "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": 4.928303923341276,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.6554,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15289047360420227,
      "step": 5,
      "valid_targets_mean": 5641.3,
      "valid_targets_min": 1301
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.56587737821047,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.6827,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13833564519882202,
      "step": 10,
      "valid_targets_mean": 5444.6,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.096,
      "grad_norm": 1.896045909613139,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.6055,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1605866253376007,
      "step": 15,
      "valid_targets_mean": 5732.6,
      "valid_targets_min": 2447
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.7667010725078626,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.5813,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13460011780261993,
      "step": 20,
      "valid_targets_mean": 5091.4,
      "valid_targets_min": 834
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6644584196631574,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.5517,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10424536466598511,
      "step": 25,
      "valid_targets_mean": 3950.8,
      "valid_targets_min": 627
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.46403980261557193,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.5304,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11924353986978531,
      "step": 30,
      "valid_targets_mean": 5288.8,
      "valid_targets_min": 695
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.39227274095601544,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5169,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15169405937194824,
      "step": 35,
      "valid_targets_mean": 5724.2,
      "valid_targets_min": 1349
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.2911708854639352,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.4812,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09291832149028778,
      "step": 40,
      "valid_targets_mean": 5563.2,
      "valid_targets_min": 1156
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.2730326283080748,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.4647,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0825074166059494,
      "step": 45,
      "valid_targets_mean": 4044.0,
      "valid_targets_min": 505
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.24544550096701181,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.4586,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08485662192106247,
      "step": 50,
      "valid_targets_mean": 4775.4,
      "valid_targets_min": 891
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.23353715280746448,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.4545,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1185971349477768,
      "step": 55,
      "valid_targets_mean": 6933.8,
      "valid_targets_min": 1072
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.2247089600498084,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.4294,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10480961948633194,
      "step": 60,
      "valid_targets_mean": 5542.1,
      "valid_targets_min": 748
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.25675788129088145,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.4524,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09944994747638702,
      "step": 65,
      "valid_targets_mean": 4435.2,
      "valid_targets_min": 666
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.24511356998076508,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.4223,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08963944017887115,
      "step": 70,
      "valid_targets_mean": 5001.8,
      "valid_targets_min": 770
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.21675938871267908,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.4355,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09129945933818817,
      "step": 75,
      "valid_targets_mean": 5227.5,
      "valid_targets_min": 442
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.2629108825099611,
      "learning_rate": 4e-05,
      "loss": 0.4259,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09954790025949478,
      "step": 80,
      "valid_targets_mean": 4703.3,
      "valid_targets_min": 712
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.22125338534597627,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.409,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09648779034614563,
      "step": 85,
      "valid_targets_mean": 6303.9,
      "valid_targets_min": 659
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.23276290152257195,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.4085,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10333982855081558,
      "step": 90,
      "valid_targets_mean": 4726.8,
      "valid_targets_min": 670
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.21517810450264546,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4112,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10249157249927521,
      "step": 95,
      "valid_targets_mean": 5971.6,
      "valid_targets_min": 974
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.25147290111354365,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.403,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10272113978862762,
      "step": 100,
      "valid_targets_mean": 5096.6,
      "valid_targets_min": 886
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.23401564883026865,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.4054,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12163965404033661,
      "step": 105,
      "valid_targets_mean": 6824.0,
      "valid_targets_min": 760
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.2290479008287171,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4004,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1165388897061348,
      "step": 110,
      "valid_targets_mean": 6983.6,
      "valid_targets_min": 1467
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.23839651513534693,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.3999,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11183565109968185,
      "step": 115,
      "valid_targets_mean": 5738.2,
      "valid_targets_min": 576
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.21718928576593474,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.3989,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10332593321800232,
      "step": 120,
      "valid_targets_mean": 5972.9,
      "valid_targets_min": 1069
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.22454969467766567,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4012,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08161451667547226,
      "step": 125,
      "valid_targets_mean": 6150.6,
      "valid_targets_min": 1019
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.2338808197665691,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.3994,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10118290781974792,
      "step": 130,
      "valid_targets_mean": 5070.9,
      "valid_targets_min": 687
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.22360556626323538,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.4046,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09752712398767471,
      "step": 135,
      "valid_targets_mean": 4816.9,
      "valid_targets_min": 496
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.23997680210076047,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.3888,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08181541413068771,
      "step": 140,
      "valid_targets_mean": 5070.4,
      "valid_targets_min": 729
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.22451344205486704,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.3733,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09583885967731476,
      "step": 145,
      "valid_targets_mean": 6154.8,
      "valid_targets_min": 812
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.26890310548364266,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.385,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11868951469659805,
      "step": 150,
      "valid_targets_mean": 5690.5,
      "valid_targets_min": 987
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.22733298879608416,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.3809,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.090506911277771,
      "step": 155,
      "valid_targets_mean": 5883.9,
      "valid_targets_min": 676
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.23726768840076212,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.392,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08799299597740173,
      "step": 160,
      "valid_targets_mean": 4297.6,
      "valid_targets_min": 575
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.22801340550293098,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.3776,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09078355133533478,
      "step": 165,
      "valid_targets_mean": 5597.7,
      "valid_targets_min": 683
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.22357443265551508,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.3844,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0776141956448555,
      "step": 170,
      "valid_targets_mean": 4806.2,
      "valid_targets_min": 858
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.22946285324328855,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.3747,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09163686633110046,
      "step": 175,
      "valid_targets_mean": 4686.6,
      "valid_targets_min": 535
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.23011603198812142,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.377,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10034259408712387,
      "step": 180,
      "valid_targets_mean": 5953.7,
      "valid_targets_min": 1237
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.25380179303397327,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.3784,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08439832180738449,
      "step": 185,
      "valid_targets_mean": 4131.6,
      "valid_targets_min": 826
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.20988638695701559,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.3802,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09300483018159866,
      "step": 190,
      "valid_targets_mean": 5937.2,
      "valid_targets_min": 735
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.22239347291739783,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.3687,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10371306538581848,
      "step": 195,
      "valid_targets_mean": 6962.8,
      "valid_targets_min": 2016
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.22763471960424564,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.3783,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09472569823265076,
      "step": 200,
      "valid_targets_mean": 5437.2,
      "valid_targets_min": 824
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.24064510945856749,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.3795,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0747733861207962,
      "step": 205,
      "valid_targets_mean": 5013.6,
      "valid_targets_min": 712
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.23942330379692994,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.3835,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11098362505435944,
      "step": 210,
      "valid_targets_mean": 6360.2,
      "valid_targets_min": 1310
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.23452470886409366,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.3764,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07871714234352112,
      "step": 215,
      "valid_targets_mean": 4264.9,
      "valid_targets_min": 544
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.23798538323500035,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.3764,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10644161701202393,
      "step": 220,
      "valid_targets_mean": 6411.9,
      "valid_targets_min": 615
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.2261263393303569,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.3711,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10277056694030762,
      "step": 225,
      "valid_targets_mean": 5812.1,
      "valid_targets_min": 747
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.2081964197146565,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.3758,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0956193208694458,
      "step": 230,
      "valid_targets_mean": 6630.9,
      "valid_targets_min": 1951
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.2572240872801356,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.374,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10277930647134781,
      "step": 235,
      "valid_targets_mean": 5665.6,
      "valid_targets_min": 946
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.2198363702134106,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.3749,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07211177796125412,
      "step": 240,
      "valid_targets_mean": 4442.8,
      "valid_targets_min": 627
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.23165428741215732,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.3733,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09441053122282028,
      "step": 245,
      "valid_targets_mean": 6673.9,
      "valid_targets_min": 1380
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.24688797409612753,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.3948,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09544113278388977,
      "step": 250,
      "valid_targets_mean": 6358.1,
      "valid_targets_min": 988
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.22463007550132932,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.3738,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07425335049629211,
      "step": 255,
      "valid_targets_mean": 5924.9,
      "valid_targets_min": 699
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.21528454307824493,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.3689,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09151971340179443,
      "step": 260,
      "valid_targets_mean": 6274.9,
      "valid_targets_min": 840
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.2215410876562211,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.3695,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08906073123216629,
      "step": 265,
      "valid_targets_mean": 4970.8,
      "valid_targets_min": 717
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.2233444272815666,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.3625,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10551338642835617,
      "step": 270,
      "valid_targets_mean": 6200.3,
      "valid_targets_min": 834
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.22527865089628543,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.3594,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09910030663013458,
      "step": 275,
      "valid_targets_mean": 5179.2,
      "valid_targets_min": 1147
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.22964617053318398,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.3706,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10371363162994385,
      "step": 280,
      "valid_targets_mean": 6714.1,
      "valid_targets_min": 1588
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.232247066108661,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.3658,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0978105291724205,
      "step": 285,
      "valid_targets_mean": 5761.8,
      "valid_targets_min": 994
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.21669192347183597,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.3701,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08539731800556183,
      "step": 290,
      "valid_targets_mean": 6083.4,
      "valid_targets_min": 672
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.22641603063104573,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.3745,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09556318819522858,
      "step": 295,
      "valid_targets_mean": 5751.5,
      "valid_targets_min": 805
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.2377563867170581,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.3669,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1070186123251915,
      "step": 300,
      "valid_targets_mean": 6680.9,
      "valid_targets_min": 718
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.22531794652697687,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.3658,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08206994831562042,
      "step": 305,
      "valid_targets_mean": 5516.9,
      "valid_targets_min": 605
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.2527878693741202,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.3767,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08087100088596344,
      "step": 310,
      "valid_targets_mean": 4204.3,
      "valid_targets_min": 663
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.3321661624480813,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.373,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09289988875389099,
      "step": 315,
      "valid_targets_mean": 6363.2,
      "valid_targets_min": 1309
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.2189394760470412,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.3706,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10759694874286652,
      "step": 320,
      "valid_targets_mean": 6793.4,
      "valid_targets_min": 842
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.24277780562479134,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.3625,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10258517414331436,
      "step": 325,
      "valid_targets_mean": 7012.6,
      "valid_targets_min": 962
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.21815774052393153,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.3534,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07705284655094147,
      "step": 330,
      "valid_targets_mean": 6053.9,
      "valid_targets_min": 631
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.23358579698798576,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.3566,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07068303227424622,
      "step": 335,
      "valid_targets_mean": 4342.3,
      "valid_targets_min": 831
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.2293951107938635,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.3581,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11601412296295166,
      "step": 340,
      "valid_targets_mean": 6150.0,
      "valid_targets_min": 798
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.22196286676468407,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.3578,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07979711145162582,
      "step": 345,
      "valid_targets_mean": 6211.4,
      "valid_targets_min": 1315
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.23510132867860487,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.3584,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1026277244091034,
      "step": 350,
      "valid_targets_mean": 6624.2,
      "valid_targets_min": 773
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.2372214737574988,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.3574,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0727909654378891,
      "step": 355,
      "valid_targets_mean": 4593.9,
      "valid_targets_min": 779
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.2608354088589743,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.3558,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08886592090129852,
      "step": 360,
      "valid_targets_mean": 5197.9,
      "valid_targets_min": 720
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.21824482003381773,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.3644,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08320988714694977,
      "step": 365,
      "valid_targets_mean": 5329.1,
      "valid_targets_min": 1164
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.21235422984856192,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.3625,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09808023273944855,
      "step": 370,
      "valid_targets_mean": 6357.5,
      "valid_targets_min": 843
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.2106692959936181,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3602,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09965196251869202,
      "step": 375,
      "valid_targets_mean": 6617.5,
      "valid_targets_min": 475
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.2304091700869004,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.3596,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0734720453619957,
      "step": 380,
      "valid_targets_mean": 4741.2,
      "valid_targets_min": 724
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.2267539238374325,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.3575,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11137191951274872,
      "step": 385,
      "valid_targets_mean": 6512.4,
      "valid_targets_min": 1490
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.22483470360256477,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.3561,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0827735960483551,
      "step": 390,
      "valid_targets_mean": 5092.9,
      "valid_targets_min": 603
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.2338040760755172,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.3614,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09697049856185913,
      "step": 395,
      "valid_targets_mean": 6565.7,
      "valid_targets_min": 1146
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.21775737564680825,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.358,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06826545298099518,
      "step": 400,
      "valid_targets_mean": 4616.6,
      "valid_targets_min": 484
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.2393125837673912,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.3638,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08487249910831451,
      "step": 405,
      "valid_targets_mean": 5329.2,
      "valid_targets_min": 589
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.22489897682687407,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.3619,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07958868145942688,
      "step": 410,
      "valid_targets_mean": 4936.8,
      "valid_targets_min": 861
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.2286817452241084,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.3605,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08212600648403168,
      "step": 415,
      "valid_targets_mean": 4606.9,
      "valid_targets_min": 615
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.22186777671536473,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.3636,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08989620208740234,
      "step": 420,
      "valid_targets_mean": 5660.1,
      "valid_targets_min": 644
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.20696912539389964,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.3633,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09872286021709442,
      "step": 425,
      "valid_targets_mean": 7214.3,
      "valid_targets_min": 920
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.2808341896352063,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.3627,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1094541922211647,
      "step": 430,
      "valid_targets_mean": 5776.4,
      "valid_targets_min": 585
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.22116930433764978,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.3491,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08739478141069412,
      "step": 435,
      "valid_targets_mean": 5458.2,
      "valid_targets_min": 619
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.1992825171622233,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3513,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09611988812685013,
      "step": 440,
      "valid_targets_mean": 7524.6,
      "valid_targets_min": 997
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.2060388593577526,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.3619,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06781405955553055,
      "step": 445,
      "valid_targets_mean": 4123.4,
      "valid_targets_min": 972
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.23527602817119048,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3566,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0758148729801178,
      "step": 450,
      "valid_targets_mean": 4773.1,
      "valid_targets_min": 535
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.22860261154505498,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.346,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09533117711544037,
      "step": 455,
      "valid_targets_mean": 4987.8,
      "valid_targets_min": 956
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.2334098694639875,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.3542,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09452962875366211,
      "step": 460,
      "valid_targets_mean": 5574.0,
      "valid_targets_min": 494
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.2126977051969965,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.3512,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08022388070821762,
      "step": 465,
      "valid_targets_mean": 6034.9,
      "valid_targets_min": 783
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.22120679343053806,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.362,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0962652713060379,
      "step": 470,
      "valid_targets_mean": 5085.6,
      "valid_targets_min": 924
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.20297765680380747,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.3577,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10209856182336807,
      "step": 475,
      "valid_targets_mean": 7576.9,
      "valid_targets_min": 2449
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.2434909996765562,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.3594,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09686852991580963,
      "step": 480,
      "valid_targets_mean": 6337.4,
      "valid_targets_min": 597
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.20464777567438086,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.3447,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08646319061517715,
      "step": 485,
      "valid_targets_mean": 6353.4,
      "valid_targets_min": 839
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.25498885027205,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.359,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10151713341474533,
      "step": 490,
      "valid_targets_mean": 6548.2,
      "valid_targets_min": 560
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.22005098708946666,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.3574,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0828465074300766,
      "step": 495,
      "valid_targets_mean": 5273.4,
      "valid_targets_min": 926
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.20776545841167376,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.3557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08530064672231674,
      "step": 500,
      "valid_targets_mean": 5513.8,
      "valid_targets_min": 761
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.2075592694245999,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3531,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09321903437376022,
      "step": 505,
      "valid_targets_mean": 6665.9,
      "valid_targets_min": 610
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.22448009174393752,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3507,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08991958945989609,
      "step": 510,
      "valid_targets_mean": 6045.4,
      "valid_targets_min": 580
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.21921091363267778,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.3525,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09069504588842392,
      "step": 515,
      "valid_targets_mean": 5085.8,
      "valid_targets_min": 724
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.20133283928290419,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3582,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10287865251302719,
      "step": 520,
      "valid_targets_mean": 7327.9,
      "valid_targets_min": 1160
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.21835206878527572,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.3409,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08309683203697205,
      "step": 525,
      "valid_targets_mean": 4695.1,
      "valid_targets_min": 617
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.20615396657777202,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3429,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08827657997608185,
      "step": 530,
      "valid_targets_mean": 6441.4,
      "valid_targets_min": 607
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.21951971461105507,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.3517,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09069855511188507,
      "step": 535,
      "valid_targets_mean": 5660.6,
      "valid_targets_min": 760
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.2155460726831645,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.3528,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0846237763762474,
      "step": 540,
      "valid_targets_mean": 7381.5,
      "valid_targets_min": 855
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.3203715725671402,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.3421,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08102697879076004,
      "step": 545,
      "valid_targets_mean": 4708.5,
      "valid_targets_min": 631
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.23643506317896415,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.352,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09795521199703217,
      "step": 550,
      "valid_targets_mean": 5957.8,
      "valid_targets_min": 1062
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.2088244497904259,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3516,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08050885051488876,
      "step": 555,
      "valid_targets_mean": 5044.3,
      "valid_targets_min": 787
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.23242692183503894,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.3581,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1041894406080246,
      "step": 560,
      "valid_targets_mean": 6474.5,
      "valid_targets_min": 544
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.22834355253798397,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.347,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08437982946634293,
      "step": 565,
      "valid_targets_mean": 5337.5,
      "valid_targets_min": 762
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.20698633953021922,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3504,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09090393036603928,
      "step": 570,
      "valid_targets_mean": 5707.0,
      "valid_targets_min": 627
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.23868065924047968,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.3501,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11131159961223602,
      "step": 575,
      "valid_targets_mean": 6219.8,
      "valid_targets_min": 1551
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.21713561340423976,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.3537,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07572489976882935,
      "step": 580,
      "valid_targets_mean": 5110.4,
      "valid_targets_min": 965
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.23258151301582297,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.3504,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09515802562236786,
      "step": 585,
      "valid_targets_mean": 5831.9,
      "valid_targets_min": 748
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.19344225479499663,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3365,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08368784189224243,
      "step": 590,
      "valid_targets_mean": 6252.3,
      "valid_targets_min": 949
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.20236324899655156,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.3523,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.077367402613163,
      "step": 595,
      "valid_targets_mean": 5667.1,
      "valid_targets_min": 1145
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.21389585445053377,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.3525,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06701831519603729,
      "step": 600,
      "valid_targets_mean": 4404.4,
      "valid_targets_min": 673
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.2141988047027784,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3474,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07725034654140472,
      "step": 605,
      "valid_targets_mean": 4666.8,
      "valid_targets_min": 672
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.21788436045315057,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.3626,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09653319418430328,
      "step": 610,
      "valid_targets_mean": 6227.8,
      "valid_targets_min": 633
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.20712703833581173,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.3465,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08920042216777802,
      "step": 615,
      "valid_targets_mean": 6852.4,
      "valid_targets_min": 841
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.24995952943643207,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.3469,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09743104875087738,
      "step": 620,
      "valid_targets_mean": 6217.0,
      "valid_targets_min": 1621
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.2159433390377831,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.3435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09399717301130295,
      "step": 625,
      "valid_targets_mean": 6170.4,
      "valid_targets_min": 996
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.2179842780291344,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3413,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08746038377285004,
      "step": 630,
      "valid_targets_mean": 5678.9,
      "valid_targets_min": 1098
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.21661284326001473,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.3484,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09619256854057312,
      "step": 635,
      "valid_targets_mean": 5427.3,
      "valid_targets_min": 610
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.21233039859449135,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.3497,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09747956693172455,
      "step": 640,
      "valid_targets_mean": 6232.0,
      "valid_targets_min": 1075
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.22408424913647904,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.3436,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0862206369638443,
      "step": 645,
      "valid_targets_mean": 5968.9,
      "valid_targets_min": 679
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.20534852988524022,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.3407,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06521781533956528,
      "step": 650,
      "valid_targets_mean": 4388.8,
      "valid_targets_min": 524
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.2239744906665863,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3464,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10012225806713104,
      "step": 655,
      "valid_targets_mean": 6456.8,
      "valid_targets_min": 1832
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.22249692849502337,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.3477,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07816874235868454,
      "step": 660,
      "valid_targets_mean": 4665.2,
      "valid_targets_min": 900
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.21634894492727133,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.35,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08766955137252808,
      "step": 665,
      "valid_targets_mean": 5389.8,
      "valid_targets_min": 689
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.21888160573451695,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3422,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07027369737625122,
      "step": 670,
      "valid_targets_mean": 4378.2,
      "valid_targets_min": 633
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.2201626385970148,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3329,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10017628222703934,
      "step": 675,
      "valid_targets_mean": 6495.4,
      "valid_targets_min": 814
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.19188368547158766,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3474,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0863613709807396,
      "step": 680,
      "valid_targets_mean": 6659.5,
      "valid_targets_min": 667
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.21926521201305976,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.3479,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07919745147228241,
      "step": 685,
      "valid_targets_mean": 4461.8,
      "valid_targets_min": 810
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.2118446211857856,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3497,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09431198239326477,
      "step": 690,
      "valid_targets_mean": 6019.9,
      "valid_targets_min": 562
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.21603812442782785,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.3498,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10296124219894409,
      "step": 695,
      "valid_targets_mean": 5586.2,
      "valid_targets_min": 697
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.21210949288960076,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.3519,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07699444890022278,
      "step": 700,
      "valid_targets_mean": 4835.4,
      "valid_targets_min": 693
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.19395198845710182,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3417,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09297066181898117,
      "step": 705,
      "valid_targets_mean": 6489.8,
      "valid_targets_min": 989
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.21037156758299613,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.3498,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10028454661369324,
      "step": 710,
      "valid_targets_mean": 6283.6,
      "valid_targets_min": 832
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.2078678569930235,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.3533,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1038275957107544,
      "step": 715,
      "valid_targets_mean": 6109.6,
      "valid_targets_min": 960
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.1913071283342328,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3428,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07966470718383789,
      "step": 720,
      "valid_targets_mean": 5774.0,
      "valid_targets_min": 1182
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.2056466084178444,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.3541,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09945742785930634,
      "step": 725,
      "valid_targets_mean": 6061.4,
      "valid_targets_min": 1024
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.20355768534921426,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.3416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07647189497947693,
      "step": 730,
      "valid_targets_mean": 5401.0,
      "valid_targets_min": 627
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.2971494219736067,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.344,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09119421243667603,
      "step": 735,
      "valid_targets_mean": 5738.6,
      "valid_targets_min": 1019
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.20448301721560436,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3447,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08554182946681976,
      "step": 740,
      "valid_targets_mean": 5514.2,
      "valid_targets_min": 719
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.19162795078979308,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.3519,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08787985146045685,
      "step": 745,
      "valid_targets_mean": 6579.6,
      "valid_targets_min": 1813
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.2000345137923447,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.3456,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0857427567243576,
      "step": 750,
      "valid_targets_mean": 5970.8,
      "valid_targets_min": 747
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.20046177963962433,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.3435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08476918935775757,
      "step": 755,
      "valid_targets_mean": 6556.9,
      "valid_targets_min": 580
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.2138261530657323,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3484,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07367642223834991,
      "step": 760,
      "valid_targets_mean": 4413.4,
      "valid_targets_min": 762
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.1963797355809849,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3561,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09871198236942291,
      "step": 765,
      "valid_targets_mean": 6162.2,
      "valid_targets_min": 1315
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.29758135156186083,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.345,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09883379936218262,
      "step": 770,
      "valid_targets_mean": 5996.9,
      "valid_targets_min": 668
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.19673112260685463,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.3453,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10000074654817581,
      "step": 775,
      "valid_targets_mean": 6430.5,
      "valid_targets_min": 1007
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.2093787903918503,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3602,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0803316980600357,
      "step": 780,
      "valid_targets_mean": 4864.9,
      "valid_targets_min": 580
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3831034791262974,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3457,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.34229332208633423,
      "step": 785,
      "valid_targets_mean": 5900.6,
      "valid_targets_min": 645
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.34229332208633423,
      "step": 785,
      "total_flos": 1.880116230107431e+18,
      "train_loss": 0.3771306716712417,
      "train_runtime": 24435.7738,
      "train_samples_per_second": 2.046,
      "train_steps_per_second": 0.032,
      "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": 1.880116230107431e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}