{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 785,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032,
      "grad_norm": 6.107608851383935,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.7455,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17644742131233215,
      "step": 5,
      "valid_targets_mean": 4706.8,
      "valid_targets_min": 1106
    },
    {
      "epoch": 0.064,
      "grad_norm": 4.127049955682705,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.7168,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15632694959640503,
      "step": 10,
      "valid_targets_mean": 4084.8,
      "valid_targets_min": 810
    },
    {
      "epoch": 0.096,
      "grad_norm": 2.3029768018854244,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.662,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1367417573928833,
      "step": 15,
      "valid_targets_mean": 4292.0,
      "valid_targets_min": 1183
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.9446288361337228,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.6259,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17443199455738068,
      "step": 20,
      "valid_targets_mean": 4366.4,
      "valid_targets_min": 794
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7185214756105344,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.6034,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15555769205093384,
      "step": 25,
      "valid_targets_mean": 4660.7,
      "valid_targets_min": 791
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.5760489151085344,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.5807,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10560796409845352,
      "step": 30,
      "valid_targets_mean": 2108.0,
      "valid_targets_min": 626
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.4877135466876705,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5433,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11195793747901917,
      "step": 35,
      "valid_targets_mean": 2955.8,
      "valid_targets_min": 986
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.3624410004020328,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.5255,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13301032781600952,
      "step": 40,
      "valid_targets_mean": 4398.3,
      "valid_targets_min": 974
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.3161733355122809,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.5254,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09797270596027374,
      "step": 45,
      "valid_targets_mean": 3789.7,
      "valid_targets_min": 971
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.3282787112321606,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.504,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11803701519966125,
      "step": 50,
      "valid_targets_mean": 3716.9,
      "valid_targets_min": 745
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.3304161820612071,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.5226,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1383068859577179,
      "step": 55,
      "valid_targets_mean": 3657.2,
      "valid_targets_min": 742
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.26548459531887825,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.4865,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1221662163734436,
      "step": 60,
      "valid_targets_mean": 4677.1,
      "valid_targets_min": 790
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.28910632104468453,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.484,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11457986384630203,
      "step": 65,
      "valid_targets_mean": 3794.5,
      "valid_targets_min": 1093
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.28039748750342514,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.4681,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13367806375026703,
      "step": 70,
      "valid_targets_mean": 4306.7,
      "valid_targets_min": 811
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.3515687541623445,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.474,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09756779670715332,
      "step": 75,
      "valid_targets_mean": 3202.6,
      "valid_targets_min": 995
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.29574743171943335,
      "learning_rate": 4e-05,
      "loss": 0.4638,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10536515712738037,
      "step": 80,
      "valid_targets_mean": 3387.2,
      "valid_targets_min": 778
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.304879438569274,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.4739,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09783598780632019,
      "step": 85,
      "valid_targets_mean": 2869.9,
      "valid_targets_min": 769
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.27264054448960384,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.449,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1292189657688141,
      "step": 90,
      "valid_targets_mean": 5253.2,
      "valid_targets_min": 1530
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.28510371417891295,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09879469871520996,
      "step": 95,
      "valid_targets_mean": 3112.0,
      "valid_targets_min": 909
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3042715840047225,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.4534,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11518902331590652,
      "step": 100,
      "valid_targets_mean": 3185.5,
      "valid_targets_min": 896
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.3122541792144832,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.4602,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10355521738529205,
      "step": 105,
      "valid_targets_mean": 3099.8,
      "valid_targets_min": 877
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.3010397462755949,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4583,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09757611155509949,
      "step": 110,
      "valid_targets_mean": 2974.1,
      "valid_targets_min": 735
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.2723793782396818,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.4453,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0976935625076294,
      "step": 115,
      "valid_targets_mean": 4154.2,
      "valid_targets_min": 932
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.2777107583881316,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.4403,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11072695255279541,
      "step": 120,
      "valid_targets_mean": 4520.2,
      "valid_targets_min": 989
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.29626977921824504,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4448,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09114716947078705,
      "step": 125,
      "valid_targets_mean": 3407.4,
      "valid_targets_min": 1066
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.27035588866639104,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.4289,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10672830790281296,
      "step": 130,
      "valid_targets_mean": 4044.8,
      "valid_targets_min": 909
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.34837201643421056,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.4476,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1184217780828476,
      "step": 135,
      "valid_targets_mean": 3964.4,
      "valid_targets_min": 633
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.27938671284433453,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.4455,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12027984112501144,
      "step": 140,
      "valid_targets_mean": 4521.9,
      "valid_targets_min": 1131
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.31804065299221124,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.445,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0931854322552681,
      "step": 145,
      "valid_targets_mean": 3013.6,
      "valid_targets_min": 766
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.2744925417793362,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.4523,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10946588218212128,
      "step": 150,
      "valid_targets_mean": 4371.4,
      "valid_targets_min": 541
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.2856537317229886,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.4414,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11907991021871567,
      "step": 155,
      "valid_targets_mean": 4329.2,
      "valid_targets_min": 624
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.2879141707463672,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.4448,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11861169338226318,
      "step": 160,
      "valid_targets_mean": 4391.5,
      "valid_targets_min": 778
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.29661655419869887,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.439,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10656130313873291,
      "step": 165,
      "valid_targets_mean": 3601.3,
      "valid_targets_min": 1213
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.28570287652110415,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.4328,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1121365949511528,
      "step": 170,
      "valid_targets_mean": 4714.2,
      "valid_targets_min": 965
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.27398298369656376,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.4357,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12218301743268967,
      "step": 175,
      "valid_targets_mean": 4598.1,
      "valid_targets_min": 1201
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.2711176693333212,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.4112,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09930449724197388,
      "step": 180,
      "valid_targets_mean": 3603.8,
      "valid_targets_min": 699
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.2929354661757715,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.4311,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10408104211091995,
      "step": 185,
      "valid_targets_mean": 3858.0,
      "valid_targets_min": 899
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.3048058460665974,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.4192,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09669551253318787,
      "step": 190,
      "valid_targets_mean": 3769.9,
      "valid_targets_min": 614
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.30457339863231014,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.4117,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0848011001944542,
      "step": 195,
      "valid_targets_mean": 3365.2,
      "valid_targets_min": 750
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.26188787177664613,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.4177,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08277255296707153,
      "step": 200,
      "valid_targets_mean": 3634.0,
      "valid_targets_min": 1159
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.28252106293682183,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.4233,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09822885692119598,
      "step": 205,
      "valid_targets_mean": 3658.9,
      "valid_targets_min": 1293
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.290327599906892,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.4194,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09229561686515808,
      "step": 210,
      "valid_targets_mean": 3470.6,
      "valid_targets_min": 936
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.2922145234243129,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.4182,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09459995478391647,
      "step": 215,
      "valid_targets_mean": 3381.3,
      "valid_targets_min": 854
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.30800560949213257,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.4275,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15284359455108643,
      "step": 220,
      "valid_targets_mean": 4698.0,
      "valid_targets_min": 1621
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.2946060468124792,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.4232,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1023549735546112,
      "step": 225,
      "valid_targets_mean": 3266.2,
      "valid_targets_min": 696
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.2909375667048486,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.4253,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10413042455911636,
      "step": 230,
      "valid_targets_mean": 3986.3,
      "valid_targets_min": 931
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.258968652511318,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.4105,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08242276310920715,
      "step": 235,
      "valid_targets_mean": 3288.0,
      "valid_targets_min": 825
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.260120898792648,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.4085,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1041165292263031,
      "step": 240,
      "valid_targets_mean": 4548.1,
      "valid_targets_min": 1250
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.3815842876872111,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.4129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10319903492927551,
      "step": 245,
      "valid_targets_mean": 4928.2,
      "valid_targets_min": 1525
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.29887518088782095,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.4116,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09942469000816345,
      "step": 250,
      "valid_targets_mean": 4006.8,
      "valid_targets_min": 923
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.3077625843249331,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.4294,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11495564132928848,
      "step": 255,
      "valid_targets_mean": 3654.4,
      "valid_targets_min": 633
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.30742900313420013,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.4269,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07939112186431885,
      "step": 260,
      "valid_targets_mean": 2838.2,
      "valid_targets_min": 651
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.28353397923235046,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.4164,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11654111742973328,
      "step": 265,
      "valid_targets_mean": 5176.1,
      "valid_targets_min": 1145
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.3159988844954844,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.412,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10565860569477081,
      "step": 270,
      "valid_targets_mean": 3831.5,
      "valid_targets_min": 885
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.28303832854596667,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.4199,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09378932416439056,
      "step": 275,
      "valid_targets_mean": 4078.2,
      "valid_targets_min": 1085
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.29629452883670093,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.4151,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11147819459438324,
      "step": 280,
      "valid_targets_mean": 4165.8,
      "valid_targets_min": 1130
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.30579345790370027,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.4165,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09240031242370605,
      "step": 285,
      "valid_targets_mean": 3160.4,
      "valid_targets_min": 964
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.2639801571915376,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.4019,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10795797407627106,
      "step": 290,
      "valid_targets_mean": 5055.9,
      "valid_targets_min": 1849
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.27607877025115213,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.4125,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12126939743757248,
      "step": 295,
      "valid_targets_mean": 5233.9,
      "valid_targets_min": 1737
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.2669974360426593,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.4035,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09347060322761536,
      "step": 300,
      "valid_targets_mean": 4767.1,
      "valid_targets_min": 1224
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.3345770824254686,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.4243,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0994449108839035,
      "step": 305,
      "valid_targets_mean": 3115.7,
      "valid_targets_min": 787
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.2848512696158886,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.4156,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09286624193191528,
      "step": 310,
      "valid_targets_mean": 3558.1,
      "valid_targets_min": 793
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.3082601924350668,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.4175,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13024066388607025,
      "step": 315,
      "valid_targets_mean": 4633.3,
      "valid_targets_min": 710
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.2972613402450226,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.408,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14002487063407898,
      "step": 320,
      "valid_targets_mean": 5598.2,
      "valid_targets_min": 1014
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.29338748624504957,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.4075,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11791073530912399,
      "step": 325,
      "valid_targets_mean": 4469.2,
      "valid_targets_min": 909
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.2788624095471663,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.4052,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09456164389848709,
      "step": 330,
      "valid_targets_mean": 2896.9,
      "valid_targets_min": 1122
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.2581699312084376,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.3929,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09569685161113739,
      "step": 335,
      "valid_targets_mean": 3911.0,
      "valid_targets_min": 940
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.31608505858012337,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.4008,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1096731573343277,
      "step": 340,
      "valid_targets_mean": 4302.9,
      "valid_targets_min": 1251
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.2880911863813072,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.4025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09002970159053802,
      "step": 345,
      "valid_targets_mean": 3883.4,
      "valid_targets_min": 841
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.33383503829503497,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.3976,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11322958767414093,
      "step": 350,
      "valid_targets_mean": 4466.8,
      "valid_targets_min": 890
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.3060080665890357,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.3962,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1092362031340599,
      "step": 355,
      "valid_targets_mean": 3826.1,
      "valid_targets_min": 1137
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.26221504446327776,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.398,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08882156014442444,
      "step": 360,
      "valid_targets_mean": 4397.3,
      "valid_targets_min": 1033
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.3340098068516625,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.4117,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12631461024284363,
      "step": 365,
      "valid_targets_mean": 4953.4,
      "valid_targets_min": 2095
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.27145668481073304,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.403,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10268014669418335,
      "step": 370,
      "valid_targets_mean": 4452.4,
      "valid_targets_min": 899
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.2810788522245723,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3973,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1246403232216835,
      "step": 375,
      "valid_targets_mean": 5544.6,
      "valid_targets_min": 1007
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.3126736568714705,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.4017,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08612629771232605,
      "step": 380,
      "valid_targets_mean": 2997.2,
      "valid_targets_min": 810
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.2953368163016378,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.4033,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08532306551933289,
      "step": 385,
      "valid_targets_mean": 3594.6,
      "valid_targets_min": 752
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.29200179241732954,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.3995,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09730486571788788,
      "step": 390,
      "valid_targets_mean": 4327.1,
      "valid_targets_min": 783
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.2946762576979655,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.4008,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11689925938844681,
      "step": 395,
      "valid_targets_mean": 4077.4,
      "valid_targets_min": 965
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.28832935456119246,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.4065,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09628384560346603,
      "step": 400,
      "valid_targets_mean": 3998.8,
      "valid_targets_min": 948
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.28622185841806863,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.409,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0933937281370163,
      "step": 405,
      "valid_targets_mean": 4487.5,
      "valid_targets_min": 801
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.2746389328095657,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.3978,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11653994023799896,
      "step": 410,
      "valid_targets_mean": 4645.2,
      "valid_targets_min": 656
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.3133949368221043,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.4034,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11006447672843933,
      "step": 415,
      "valid_targets_mean": 3497.1,
      "valid_targets_min": 1033
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.26840721362327946,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.4014,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10718810558319092,
      "step": 420,
      "valid_targets_mean": 4133.1,
      "valid_targets_min": 1486
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.28328827868904544,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.4103,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12090544402599335,
      "step": 425,
      "valid_targets_mean": 5332.2,
      "valid_targets_min": 923
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.3169828962163063,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.4139,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09149235486984253,
      "step": 430,
      "valid_targets_mean": 3602.7,
      "valid_targets_min": 827
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.24585141329828905,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.3831,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0855756402015686,
      "step": 435,
      "valid_targets_mean": 3692.9,
      "valid_targets_min": 1508
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.28624685923515464,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3972,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10186800360679626,
      "step": 440,
      "valid_targets_mean": 4585.4,
      "valid_targets_min": 877
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.2588341586751398,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.387,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07834403216838837,
      "step": 445,
      "valid_targets_mean": 3711.6,
      "valid_targets_min": 746
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.30323288737998944,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3901,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09488328546285629,
      "step": 450,
      "valid_targets_mean": 2979.2,
      "valid_targets_min": 722
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.24266113449716165,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.385,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07156480848789215,
      "step": 455,
      "valid_targets_mean": 4107.6,
      "valid_targets_min": 742
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.28261298997066275,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.393,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09054580330848694,
      "step": 460,
      "valid_targets_mean": 3750.1,
      "valid_targets_min": 1213
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.259989135927338,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.3972,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09447605907917023,
      "step": 465,
      "valid_targets_mean": 4066.9,
      "valid_targets_min": 797
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.2842931846792312,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.3914,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10005789995193481,
      "step": 470,
      "valid_targets_mean": 4615.3,
      "valid_targets_min": 1184
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.2942258678514413,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.3797,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.106219083070755,
      "step": 475,
      "valid_targets_mean": 4277.8,
      "valid_targets_min": 1094
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.2690806364664981,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.3778,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09493634104728699,
      "step": 480,
      "valid_targets_mean": 3908.0,
      "valid_targets_min": 784
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.26353669494610443,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.3819,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11369521915912628,
      "step": 485,
      "valid_targets_mean": 4864.3,
      "valid_targets_min": 972
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.4041290592713129,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.3906,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09258768707513809,
      "step": 490,
      "valid_targets_mean": 2708.4,
      "valid_targets_min": 989
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.25878391245076104,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.3897,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07488387823104858,
      "step": 495,
      "valid_targets_mean": 3082.7,
      "valid_targets_min": 656
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.253759150529026,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.3974,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1091025173664093,
      "step": 500,
      "valid_targets_mean": 4998.2,
      "valid_targets_min": 1150
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.2683080146842537,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3858,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07230888307094574,
      "step": 505,
      "valid_targets_mean": 3185.6,
      "valid_targets_min": 827
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.2947819206644499,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3903,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11434278637170792,
      "step": 510,
      "valid_targets_mean": 4241.5,
      "valid_targets_min": 734
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.2705047308271746,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.3861,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11061400175094604,
      "step": 515,
      "valid_targets_mean": 4625.5,
      "valid_targets_min": 965
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.2610230513683317,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3945,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10025350749492645,
      "step": 520,
      "valid_targets_mean": 4237.3,
      "valid_targets_min": 1048
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.2825954817857534,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.3951,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10196895897388458,
      "step": 525,
      "valid_targets_mean": 4061.5,
      "valid_targets_min": 810
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.24013472756026968,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3829,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08009563386440277,
      "step": 530,
      "valid_targets_mean": 3982.6,
      "valid_targets_min": 843
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.26036428326503036,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.4,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11866132915019989,
      "step": 535,
      "valid_targets_mean": 4450.6,
      "valid_targets_min": 1048
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.24881198574724328,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.3952,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09163567423820496,
      "step": 540,
      "valid_targets_mean": 4590.5,
      "valid_targets_min": 702
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.2843932556205226,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.3903,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09883452206850052,
      "step": 545,
      "valid_targets_mean": 3780.2,
      "valid_targets_min": 975
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.27338503039297785,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.3785,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08160857856273651,
      "step": 550,
      "valid_targets_mean": 3942.5,
      "valid_targets_min": 938
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.2674094532197557,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3901,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10117650032043457,
      "step": 555,
      "valid_targets_mean": 3955.4,
      "valid_targets_min": 793
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.27380161269817377,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.3911,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.087982177734375,
      "step": 560,
      "valid_targets_mean": 3488.1,
      "valid_targets_min": 874
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.31002515430693867,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.3873,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11359541863203049,
      "step": 565,
      "valid_targets_mean": 3674.3,
      "valid_targets_min": 941
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.32806171635804243,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3994,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08924146741628647,
      "step": 570,
      "valid_targets_mean": 2419.3,
      "valid_targets_min": 639
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.2907755504334934,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.388,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10015973448753357,
      "step": 575,
      "valid_targets_mean": 3822.8,
      "valid_targets_min": 975
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.28606169446008295,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.3811,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1018945500254631,
      "step": 580,
      "valid_targets_mean": 3379.8,
      "valid_targets_min": 612
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.2590625610172514,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.3869,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11175013333559036,
      "step": 585,
      "valid_targets_mean": 4867.4,
      "valid_targets_min": 704
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.2763907408247419,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3931,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10954703390598297,
      "step": 590,
      "valid_targets_mean": 4537.1,
      "valid_targets_min": 710
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.24962648496447917,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.3869,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09462133795022964,
      "step": 595,
      "valid_targets_mean": 4469.4,
      "valid_targets_min": 690
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.2754736478883349,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.3867,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11113976687192917,
      "step": 600,
      "valid_targets_mean": 4153.2,
      "valid_targets_min": 896
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.28158413256335013,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3822,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11500473320484161,
      "step": 605,
      "valid_targets_mean": 4267.9,
      "valid_targets_min": 863
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.2995558635935276,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.3958,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10843653976917267,
      "step": 610,
      "valid_targets_mean": 3483.2,
      "valid_targets_min": 1058
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.26971516213282304,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.4026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09091812372207642,
      "step": 615,
      "valid_targets_mean": 3210.1,
      "valid_targets_min": 693
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.27718460735351713,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.3881,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10660787671804428,
      "step": 620,
      "valid_targets_mean": 3532.4,
      "valid_targets_min": 701
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.24222704520402027,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.3854,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06686671078205109,
      "step": 625,
      "valid_targets_mean": 3113.8,
      "valid_targets_min": 1014
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.2577795312253048,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3935,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10310535132884979,
      "step": 630,
      "valid_targets_mean": 4274.4,
      "valid_targets_min": 821
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.26450694499499056,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.3818,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08196067810058594,
      "step": 635,
      "valid_targets_mean": 3661.9,
      "valid_targets_min": 798
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.25757656679535085,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.3879,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11379433423280716,
      "step": 640,
      "valid_targets_mean": 5251.4,
      "valid_targets_min": 873
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.2857282667440211,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.3898,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09233619272708893,
      "step": 645,
      "valid_targets_mean": 3476.4,
      "valid_targets_min": 942
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.26803241687834123,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.3895,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11338621377944946,
      "step": 650,
      "valid_targets_mean": 4720.1,
      "valid_targets_min": 1226
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.25400284057837186,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3711,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08650071918964386,
      "step": 655,
      "valid_targets_mean": 4031.5,
      "valid_targets_min": 809
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.2568237062355446,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.391,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09939530491828918,
      "step": 660,
      "valid_targets_mean": 4138.6,
      "valid_targets_min": 762
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.2676124398757461,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.3869,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09458170086145401,
      "step": 665,
      "valid_targets_mean": 4247.1,
      "valid_targets_min": 868
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.28046598824382457,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3807,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10842232406139374,
      "step": 670,
      "valid_targets_mean": 3938.6,
      "valid_targets_min": 833
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.2570693432350438,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3783,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08588958531618118,
      "step": 675,
      "valid_targets_mean": 3437.5,
      "valid_targets_min": 850
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.25507056895604874,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3837,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09940114617347717,
      "step": 680,
      "valid_targets_mean": 4383.1,
      "valid_targets_min": 900
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.2489526579222752,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.3837,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11270736902952194,
      "step": 685,
      "valid_targets_mean": 5063.6,
      "valid_targets_min": 960
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.30376368802757764,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3943,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08725392818450928,
      "step": 690,
      "valid_targets_mean": 3286.1,
      "valid_targets_min": 892
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.4292351264693741,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.391,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11258338391780853,
      "step": 695,
      "valid_targets_mean": 4237.6,
      "valid_targets_min": 1186
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.25325658381922306,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.3834,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0918629914522171,
      "step": 700,
      "valid_targets_mean": 4016.9,
      "valid_targets_min": 970
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.30072408255586425,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3818,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09678243845701218,
      "step": 705,
      "valid_targets_mean": 3707.1,
      "valid_targets_min": 1103
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.25410126385994386,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.3832,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10050881654024124,
      "step": 710,
      "valid_targets_mean": 4388.2,
      "valid_targets_min": 880
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.34950980807629306,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.3928,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0971115306019783,
      "step": 715,
      "valid_targets_mean": 2979.0,
      "valid_targets_min": 994
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.24312281541870698,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3839,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09972332417964935,
      "step": 720,
      "valid_targets_mean": 4491.6,
      "valid_targets_min": 969
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.2714481941303094,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.3751,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08378778398036957,
      "step": 725,
      "valid_targets_mean": 4230.4,
      "valid_targets_min": 795
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.2854227918980573,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.3818,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10359008610248566,
      "step": 730,
      "valid_targets_mean": 3959.3,
      "valid_targets_min": 1324
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.24942374686892158,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.3846,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11520683765411377,
      "step": 735,
      "valid_targets_mean": 5002.9,
      "valid_targets_min": 913
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.31797210014974603,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3957,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10495293140411377,
      "step": 740,
      "valid_targets_mean": 3320.9,
      "valid_targets_min": 706
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.2768908745456099,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.3858,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07093711942434311,
      "step": 745,
      "valid_targets_mean": 2778.2,
      "valid_targets_min": 633
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.2960407866353702,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.3758,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10425875335931778,
      "step": 750,
      "valid_targets_mean": 3789.5,
      "valid_targets_min": 896
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.2399975960595749,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.38,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09158598631620407,
      "step": 755,
      "valid_targets_mean": 4844.2,
      "valid_targets_min": 1125
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.26684033395373213,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3926,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09115723520517349,
      "step": 760,
      "valid_targets_mean": 3444.8,
      "valid_targets_min": 820
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.258854589815443,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3866,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10301363468170166,
      "step": 765,
      "valid_targets_mean": 4948.0,
      "valid_targets_min": 807
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.2624455783886108,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.3764,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08958660066127777,
      "step": 770,
      "valid_targets_mean": 3562.5,
      "valid_targets_min": 707
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.276084245733527,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.3897,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09636662155389786,
      "step": 775,
      "valid_targets_mean": 4382.5,
      "valid_targets_min": 754
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.24631738908048942,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3905,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08279719948768616,
      "step": 780,
      "valid_targets_mean": 3800.4,
      "valid_targets_min": 950
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5798931353670957,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3821,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.38632673025131226,
      "step": 785,
      "valid_targets_mean": 3155.6,
      "valid_targets_min": 917
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.38632673025131226,
      "step": 785,
      "total_flos": 1.2145665807220736e+18,
      "train_loss": 0.4196553195358082,
      "train_runtime": 16441.9592,
      "train_samples_per_second": 3.04,
      "train_steps_per_second": 0.048,
      "valid_targets_mean": 3155.6,
      "valid_targets_min": 917
    }
  ],
  "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": 1.2145665807220736e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}