{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 735,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.047619047619047616,
      "grad_norm": 3.641369581222534,
      "learning_rate": 2.1621621621621623e-06,
      "loss": 0.7903,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.7951903343200684,
      "step": 5,
      "valid_targets_mean": 9678.3,
      "valid_targets_min": 1230
    },
    {
      "epoch": 0.09523809523809523,
      "grad_norm": 3.213757276535034,
      "learning_rate": 4.864864864864866e-06,
      "loss": 0.792,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.7918487191200256,
      "step": 10,
      "valid_targets_mean": 10739.4,
      "valid_targets_min": 1620
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 1.9532331228256226,
      "learning_rate": 7.567567567567569e-06,
      "loss": 0.7561,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.737534761428833,
      "step": 15,
      "valid_targets_mean": 9632.6,
      "valid_targets_min": 997
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 1.5747649669647217,
      "learning_rate": 1.027027027027027e-05,
      "loss": 0.7117,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.6997154355049133,
      "step": 20,
      "valid_targets_mean": 11396.1,
      "valid_targets_min": 1117
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 1.5162792205810547,
      "learning_rate": 1.2972972972972975e-05,
      "loss": 0.7102,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.7041388154029846,
      "step": 25,
      "valid_targets_mean": 11505.9,
      "valid_targets_min": 1709
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.9182464480400085,
      "learning_rate": 1.5675675675675676e-05,
      "loss": 0.6922,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.6665891408920288,
      "step": 30,
      "valid_targets_mean": 10400.4,
      "valid_targets_min": 1638
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.7876750230789185,
      "learning_rate": 1.8378378378378383e-05,
      "loss": 0.6674,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.6582701206207275,
      "step": 35,
      "valid_targets_mean": 9625.1,
      "valid_targets_min": 1535
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 0.635549008846283,
      "learning_rate": 2.1081081081081082e-05,
      "loss": 0.6409,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.6444888114929199,
      "step": 40,
      "valid_targets_mean": 9969.0,
      "valid_targets_min": 1848
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 0.526866614818573,
      "learning_rate": 2.378378378378379e-05,
      "loss": 0.6243,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.6020426154136658,
      "step": 45,
      "valid_targets_mean": 9944.5,
      "valid_targets_min": 1906
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 0.5022098422050476,
      "learning_rate": 2.6486486486486488e-05,
      "loss": 0.6094,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5953571796417236,
      "step": 50,
      "valid_targets_mean": 9082.6,
      "valid_targets_min": 310
    },
    {
      "epoch": 0.5238095238095238,
      "grad_norm": 0.42319613695144653,
      "learning_rate": 2.918918918918919e-05,
      "loss": 0.595,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5980594158172607,
      "step": 55,
      "valid_targets_mean": 10010.7,
      "valid_targets_min": 905
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.3875083327293396,
      "learning_rate": 3.1891891891891894e-05,
      "loss": 0.5787,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5710088610649109,
      "step": 60,
      "valid_targets_mean": 10520.5,
      "valid_targets_min": 1355
    },
    {
      "epoch": 0.6190476190476191,
      "grad_norm": 0.32117581367492676,
      "learning_rate": 3.45945945945946e-05,
      "loss": 0.57,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5607152581214905,
      "step": 65,
      "valid_targets_mean": 10454.9,
      "valid_targets_min": 1242
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.29642170667648315,
      "learning_rate": 3.72972972972973e-05,
      "loss": 0.5496,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5441405773162842,
      "step": 70,
      "valid_targets_mean": 10958.2,
      "valid_targets_min": 1521
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.3015699088573456,
      "learning_rate": 4e-05,
      "loss": 0.5439,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5417560338973999,
      "step": 75,
      "valid_targets_mean": 11049.0,
      "valid_targets_min": 2135
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 0.3483181595802307,
      "learning_rate": 3.999435301808432e-05,
      "loss": 0.5422,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5343438982963562,
      "step": 80,
      "valid_targets_mean": 10688.3,
      "valid_targets_min": 1381
    },
    {
      "epoch": 0.8095238095238095,
      "grad_norm": 0.32726579904556274,
      "learning_rate": 3.997741526117775e-05,
      "loss": 0.5377,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5393884778022766,
      "step": 85,
      "valid_targets_mean": 9839.4,
      "valid_targets_min": 1600
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.3602434992790222,
      "learning_rate": 3.994919629400098e-05,
      "loss": 0.5262,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5330803990364075,
      "step": 90,
      "valid_targets_mean": 10885.7,
      "valid_targets_min": 1392
    },
    {
      "epoch": 0.9047619047619048,
      "grad_norm": 0.3332975506782532,
      "learning_rate": 3.990971205175375e-05,
      "loss": 0.5127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5026482939720154,
      "step": 95,
      "valid_targets_mean": 10199.3,
      "valid_targets_min": 1766
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 0.3280484080314636,
      "learning_rate": 3.985898483111624e-05,
      "loss": 0.5193,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5103785395622253,
      "step": 100,
      "valid_targets_mean": 10481.4,
      "valid_targets_min": 1970
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.2986338138580322,
      "learning_rate": 3.979704327765823e-05,
      "loss": 0.5157,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5190891027450562,
      "step": 105,
      "valid_targets_mean": 11339.4,
      "valid_targets_min": 1956
    },
    {
      "epoch": 1.0476190476190477,
      "grad_norm": 0.3262462913990021,
      "learning_rate": 3.972392236966291e-05,
      "loss": 0.5043,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5045145153999329,
      "step": 110,
      "valid_targets_mean": 10657.0,
      "valid_targets_min": 1060
    },
    {
      "epoch": 1.0952380952380953,
      "grad_norm": 0.27068766951560974,
      "learning_rate": 3.963966339837482e-05,
      "loss": 0.5053,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4862871468067169,
      "step": 115,
      "valid_targets_mean": 8995.0,
      "valid_targets_min": 1323
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.2911800146102905,
      "learning_rate": 3.954431394468266e-05,
      "loss": 0.5014,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5043173432350159,
      "step": 120,
      "valid_targets_mean": 10641.8,
      "valid_targets_min": 1669
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 0.2995125353336334,
      "learning_rate": 3.943792785225049e-05,
      "loss": 0.5028,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4877793490886688,
      "step": 125,
      "valid_targets_mean": 9839.4,
      "valid_targets_min": 2114
    },
    {
      "epoch": 1.2380952380952381,
      "grad_norm": 0.31261351704597473,
      "learning_rate": 3.932056519711232e-05,
      "loss": 0.5114,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.5172396302223206,
      "step": 130,
      "valid_targets_mean": 11658.8,
      "valid_targets_min": 1407
    },
    {
      "epoch": 1.2857142857142856,
      "grad_norm": 0.2696959674358368,
      "learning_rate": 3.919229225374726e-05,
      "loss": 0.4957,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4965190589427948,
      "step": 135,
      "valid_targets_mean": 10970.5,
      "valid_targets_min": 1465
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.31054753065109253,
      "learning_rate": 3.9053181457654465e-05,
      "loss": 0.5016,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4997459948062897,
      "step": 140,
      "valid_targets_mean": 10606.4,
      "valid_targets_min": 1938
    },
    {
      "epoch": 1.380952380952381,
      "grad_norm": 0.27902498841285706,
      "learning_rate": 3.89033113644489e-05,
      "loss": 0.4957,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.49149760603904724,
      "step": 145,
      "valid_targets_mean": 11533.9,
      "valid_targets_min": 1230
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.25977635383605957,
      "learning_rate": 3.874276660550119e-05,
      "loss": 0.4885,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4768286645412445,
      "step": 150,
      "valid_targets_mean": 10470.1,
      "valid_targets_min": 905
    },
    {
      "epoch": 1.4761904761904763,
      "grad_norm": 0.24554093182086945,
      "learning_rate": 3.857163784014636e-05,
      "loss": 0.4936,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.49420714378356934,
      "step": 155,
      "valid_targets_mean": 10378.7,
      "valid_targets_min": 1432
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 0.2620711326599121,
      "learning_rate": 3.8390021704488735e-05,
      "loss": 0.4908,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.49184074997901917,
      "step": 160,
      "valid_targets_mean": 12105.0,
      "valid_targets_min": 2425
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 0.2703709602355957,
      "learning_rate": 3.8198020756831694e-05,
      "loss": 0.4914,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4979839026927948,
      "step": 165,
      "valid_targets_mean": 10409.2,
      "valid_targets_min": 1634
    },
    {
      "epoch": 1.619047619047619,
      "grad_norm": 0.2900194525718689,
      "learning_rate": 3.799574341976314e-05,
      "loss": 0.4857,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.48033514618873596,
      "step": 170,
      "valid_targets_mean": 11115.3,
      "valid_targets_min": 1833
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.34530168771743774,
      "learning_rate": 3.778330391892952e-05,
      "loss": 0.4827,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45976319909095764,
      "step": 175,
      "valid_targets_mean": 9439.5,
      "valid_targets_min": 1920
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.2563239634037018,
      "learning_rate": 3.7560822218532774e-05,
      "loss": 0.4789,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4729023277759552,
      "step": 180,
      "valid_targets_mean": 11444.8,
      "valid_targets_min": 1647
    },
    {
      "epoch": 1.7619047619047619,
      "grad_norm": 0.2907782196998596,
      "learning_rate": 3.732842395358677e-05,
      "loss": 0.4805,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4829181730747223,
      "step": 185,
      "valid_targets_mean": 9770.2,
      "valid_targets_min": 1347
    },
    {
      "epoch": 1.8095238095238095,
      "grad_norm": 0.22722934186458588,
      "learning_rate": 3.708624035897144e-05,
      "loss": 0.4815,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4639171361923218,
      "step": 190,
      "valid_targets_mean": 10383.2,
      "valid_targets_min": 1135
    },
    {
      "epoch": 1.8571428571428572,
      "grad_norm": 0.284527063369751,
      "learning_rate": 3.68344081953247e-05,
      "loss": 0.4742,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.48767605423927307,
      "step": 195,
      "valid_targets_mean": 10920.7,
      "valid_targets_min": 1687
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.23794886469841003,
      "learning_rate": 3.657306967181394e-05,
      "loss": 0.4801,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4723459482192993,
      "step": 200,
      "valid_targets_mean": 10046.9,
      "valid_targets_min": 309
    },
    {
      "epoch": 1.9523809523809523,
      "grad_norm": 0.22136229276657104,
      "learning_rate": 3.630237236583077e-05,
      "loss": 0.479,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4783952534198761,
      "step": 205,
      "valid_targets_mean": 10083.6,
      "valid_targets_min": 1075
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.24765542149543762,
      "learning_rate": 3.6022469139654345e-05,
      "loss": 0.4772,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45786723494529724,
      "step": 210,
      "valid_targets_mean": 9495.5,
      "valid_targets_min": 2436
    },
    {
      "epoch": 2.0476190476190474,
      "grad_norm": 0.3153262734413147,
      "learning_rate": 3.57335180541303e-05,
      "loss": 0.4789,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4727652072906494,
      "step": 215,
      "valid_targets_mean": 10444.5,
      "valid_targets_min": 1230
    },
    {
      "epoch": 2.0952380952380953,
      "grad_norm": 0.2903777062892914,
      "learning_rate": 3.543568227941408e-05,
      "loss": 0.4667,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.47414258122444153,
      "step": 220,
      "valid_targets_mean": 10215.9,
      "valid_targets_min": 2024
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 0.24170319736003876,
      "learning_rate": 3.512913000282905e-05,
      "loss": 0.4748,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.46666231751441956,
      "step": 225,
      "valid_targets_mean": 10772.6,
      "valid_targets_min": 905
    },
    {
      "epoch": 2.1904761904761907,
      "grad_norm": 0.24586112797260284,
      "learning_rate": 3.481403433389142e-05,
      "loss": 0.4598,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4405348300933838,
      "step": 230,
      "valid_targets_mean": 9798.6,
      "valid_targets_min": 1521
    },
    {
      "epoch": 2.238095238095238,
      "grad_norm": 0.22028249502182007,
      "learning_rate": 3.449057320655561e-05,
      "loss": 0.4688,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45091816782951355,
      "step": 235,
      "valid_targets_mean": 10118.6,
      "valid_targets_min": 399
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.22291536629199982,
      "learning_rate": 3.415892927873527e-05,
      "loss": 0.4573,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4677928388118744,
      "step": 240,
      "valid_targets_mean": 11213.6,
      "valid_targets_min": 1075
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.21577122807502747,
      "learning_rate": 3.381928982915668e-05,
      "loss": 0.4637,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.457874059677124,
      "step": 245,
      "valid_targets_mean": 11213.5,
      "valid_targets_min": 1631
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.24737918376922607,
      "learning_rate": 3.3471846651602815e-05,
      "loss": 0.4668,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45833051204681396,
      "step": 250,
      "valid_targets_mean": 9896.9,
      "valid_targets_min": 2060
    },
    {
      "epoch": 2.4285714285714284,
      "grad_norm": 0.2504478991031647,
      "learning_rate": 3.31167959466077e-05,
      "loss": 0.4772,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45179757475852966,
      "step": 255,
      "valid_targets_mean": 9622.8,
      "valid_targets_min": 1086
    },
    {
      "epoch": 2.4761904761904763,
      "grad_norm": 0.2547358572483063,
      "learning_rate": 3.275433821066237e-05,
      "loss": 0.4673,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.46238890290260315,
      "step": 260,
      "valid_targets_mean": 10618.7,
      "valid_targets_min": 2097
    },
    {
      "epoch": 2.5238095238095237,
      "grad_norm": 0.23798689246177673,
      "learning_rate": 3.238467812299483e-05,
      "loss": 0.461,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4570256173610687,
      "step": 265,
      "valid_targets_mean": 11096.1,
      "valid_targets_min": 995
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 0.26095888018608093,
      "learning_rate": 3.200802442998807e-05,
      "loss": 0.4723,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.46601390838623047,
      "step": 270,
      "valid_targets_mean": 10761.1,
      "valid_targets_min": 1135
    },
    {
      "epoch": 2.619047619047619,
      "grad_norm": 0.27441200613975525,
      "learning_rate": 3.1624589827301395e-05,
      "loss": 0.4635,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45723530650138855,
      "step": 275,
      "valid_targets_mean": 10371.2,
      "valid_targets_min": 1285
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.23007221519947052,
      "learning_rate": 3.123459083976152e-05,
      "loss": 0.4579,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45222917199134827,
      "step": 280,
      "valid_targets_mean": 10307.8,
      "valid_targets_min": 2165
    },
    {
      "epoch": 2.7142857142857144,
      "grad_norm": 0.23393280804157257,
      "learning_rate": 3.083824769909142e-05,
      "loss": 0.4574,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44701674580574036,
      "step": 285,
      "valid_targets_mean": 9147.3,
      "valid_targets_min": 1563
    },
    {
      "epoch": 2.761904761904762,
      "grad_norm": 0.22315353155136108,
      "learning_rate": 3.0435784219545872e-05,
      "loss": 0.4529,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4526662826538086,
      "step": 290,
      "valid_targets_mean": 11007.2,
      "valid_targets_min": 2262
    },
    {
      "epoch": 2.8095238095238093,
      "grad_norm": 0.23970670998096466,
      "learning_rate": 3.0027427671523957e-05,
      "loss": 0.4562,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44976282119750977,
      "step": 295,
      "valid_targets_mean": 10453.5,
      "valid_targets_min": 1755
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.20473124086856842,
      "learning_rate": 2.961340865322984e-05,
      "loss": 0.4609,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45694804191589355,
      "step": 300,
      "valid_targets_mean": 10648.6,
      "valid_targets_min": 346
    },
    {
      "epoch": 2.9047619047619047,
      "grad_norm": 0.2585085332393646,
      "learning_rate": 2.9193960960454446e-05,
      "loss": 0.4622,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4737027585506439,
      "step": 305,
      "valid_targets_mean": 10640.5,
      "valid_targets_min": 2157
    },
    {
      "epoch": 2.9523809523809526,
      "grad_norm": 0.2357761263847351,
      "learning_rate": 2.8769321454551327e-05,
      "loss": 0.4714,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.47929224371910095,
      "step": 310,
      "valid_targets_mean": 11621.3,
      "valid_targets_min": 1407
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.22410333156585693,
      "learning_rate": 2.833972992868154e-05,
      "loss": 0.4576,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4687689244747162,
      "step": 315,
      "valid_targets_mean": 10962.5,
      "valid_targets_min": 1683
    },
    {
      "epoch": 3.0476190476190474,
      "grad_norm": 0.2000991851091385,
      "learning_rate": 2.7905428972402872e-05,
      "loss": 0.4578,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.466886043548584,
      "step": 320,
      "valid_targets_mean": 11016.1,
      "valid_targets_min": 1228
    },
    {
      "epoch": 3.0952380952380953,
      "grad_norm": 0.231069415807724,
      "learning_rate": 2.7466663834679905e-05,
      "loss": 0.4539,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4406701624393463,
      "step": 325,
      "valid_targets_mean": 9873.8,
      "valid_targets_min": 2059
    },
    {
      "epoch": 3.142857142857143,
      "grad_norm": 0.24467022716999054,
      "learning_rate": 2.7023682285392445e-05,
      "loss": 0.4529,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45116034150123596,
      "step": 330,
      "valid_targets_mean": 10860.0,
      "valid_targets_min": 1929
    },
    {
      "epoch": 3.1904761904761907,
      "grad_norm": 0.23421397805213928,
      "learning_rate": 2.657673447542028e-05,
      "loss": 0.4495,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4534952938556671,
      "step": 335,
      "valid_targets_mean": 10914.8,
      "valid_targets_min": 1075
    },
    {
      "epoch": 3.238095238095238,
      "grad_norm": 0.20964963734149933,
      "learning_rate": 2.6126072795383416e-05,
      "loss": 0.4549,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4569459855556488,
      "step": 340,
      "valid_targets_mean": 10480.9,
      "valid_targets_min": 1323
    },
    {
      "epoch": 3.2857142857142856,
      "grad_norm": 0.2521587312221527,
      "learning_rate": 2.5671951733117587e-05,
      "loss": 0.4528,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4788999557495117,
      "step": 345,
      "valid_targets_mean": 10849.1,
      "valid_targets_min": 1193
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.22168569266796112,
      "learning_rate": 2.5214627729965396e-05,
      "loss": 0.4477,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4305855333805084,
      "step": 350,
      "valid_targets_mean": 9888.9,
      "valid_targets_min": 1527
    },
    {
      "epoch": 3.380952380952381,
      "grad_norm": 0.20934681594371796,
      "learning_rate": 2.47543590359644e-05,
      "loss": 0.4518,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.451143354177475,
      "step": 355,
      "valid_targets_mean": 10741.3,
      "valid_targets_min": 1722
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 0.21030135452747345,
      "learning_rate": 2.4291405564013727e-05,
      "loss": 0.4532,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4656308591365814,
      "step": 360,
      "valid_targets_mean": 10297.5,
      "valid_targets_min": 1197
    },
    {
      "epoch": 3.4761904761904763,
      "grad_norm": 0.21608564257621765,
      "learning_rate": 2.3826028743101763e-05,
      "loss": 0.4563,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4556739330291748,
      "step": 365,
      "valid_targets_mean": 9781.0,
      "valid_targets_min": 1440
    },
    {
      "epoch": 3.5238095238095237,
      "grad_norm": 0.21990816295146942,
      "learning_rate": 2.3358491370677693e-05,
      "loss": 0.4503,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.43988335132598877,
      "step": 370,
      "valid_targets_mean": 9350.0,
      "valid_targets_min": 2125
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 0.2016260176897049,
      "learning_rate": 2.2889057464250196e-05,
      "loss": 0.4465,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4483100175857544,
      "step": 375,
      "valid_targets_mean": 9666.7,
      "valid_targets_min": 971
    },
    {
      "epoch": 3.619047619047619,
      "grad_norm": 0.2452867329120636,
      "learning_rate": 2.2417992112297293e-05,
      "loss": 0.4443,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44237199425697327,
      "step": 380,
      "valid_targets_mean": 10183.1,
      "valid_targets_min": 1600
    },
    {
      "epoch": 3.6666666666666665,
      "grad_norm": 0.22541393339633942,
      "learning_rate": 2.1945561324571366e-05,
      "loss": 0.4477,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4404517114162445,
      "step": 385,
      "valid_targets_mean": 10663.6,
      "valid_targets_min": 1391
    },
    {
      "epoch": 3.7142857142857144,
      "grad_norm": 0.21644438803195953,
      "learning_rate": 2.1472031881883856e-05,
      "loss": 0.4402,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4442715346813202,
      "step": 390,
      "valid_targets_mean": 10567.3,
      "valid_targets_min": 2345
    },
    {
      "epoch": 3.761904761904762,
      "grad_norm": 0.23878459632396698,
      "learning_rate": 2.0997671185454714e-05,
      "loss": 0.4472,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44484564661979675,
      "step": 395,
      "valid_targets_mean": 9077.4,
      "valid_targets_min": 1314
    },
    {
      "epoch": 3.8095238095238093,
      "grad_norm": 0.21471506357192993,
      "learning_rate": 2.0522747105911378e-05,
      "loss": 0.4532,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45848822593688965,
      "step": 400,
      "valid_targets_mean": 10683.8,
      "valid_targets_min": 1242
    },
    {
      "epoch": 3.857142857142857,
      "grad_norm": 0.21900102496147156,
      "learning_rate": 2.0047527832022674e-05,
      "loss": 0.4456,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4426546096801758,
      "step": 405,
      "valid_targets_mean": 9082.0,
      "valid_targets_min": 1242
    },
    {
      "epoch": 3.9047619047619047,
      "grad_norm": 0.23612160980701447,
      "learning_rate": 1.9572281719253186e-05,
      "loss": 0.4511,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.42625102400779724,
      "step": 410,
      "valid_targets_mean": 9350.7,
      "valid_targets_min": 1397
    },
    {
      "epoch": 3.9523809523809526,
      "grad_norm": 0.2100141942501068,
      "learning_rate": 1.909727713822333e-05,
      "loss": 0.455,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4657195508480072,
      "step": 415,
      "valid_targets_mean": 11093.6,
      "valid_targets_min": 1355
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.21113039553165436,
      "learning_rate": 1.8622782323161014e-05,
      "loss": 0.4548,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4469023048877716,
      "step": 420,
      "valid_targets_mean": 10532.8,
      "valid_targets_min": 2388
    },
    {
      "epoch": 4.0476190476190474,
      "grad_norm": 0.2279236912727356,
      "learning_rate": 1.8149065220430197e-05,
      "loss": 0.4419,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44380679726600647,
      "step": 425,
      "valid_targets_mean": 10220.7,
      "valid_targets_min": 915
    },
    {
      "epoch": 4.095238095238095,
      "grad_norm": 0.19087187945842743,
      "learning_rate": 1.7676393337222115e-05,
      "loss": 0.439,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4420897960662842,
      "step": 430,
      "valid_targets_mean": 10200.5,
      "valid_targets_min": 1391
    },
    {
      "epoch": 4.142857142857143,
      "grad_norm": 0.19136783480644226,
      "learning_rate": 1.7205033590494426e-05,
      "loss": 0.4396,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4429967701435089,
      "step": 435,
      "valid_targets_mean": 11265.4,
      "valid_targets_min": 1386
    },
    {
      "epoch": 4.190476190476191,
      "grad_norm": 0.21110598742961884,
      "learning_rate": 1.6735252156243675e-05,
      "loss": 0.4432,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.43789541721343994,
      "step": 440,
      "valid_targets_mean": 10057.8,
      "valid_targets_min": 1546
    },
    {
      "epoch": 4.238095238095238,
      "grad_norm": 0.20755375921726227,
      "learning_rate": 1.6267314319196215e-05,
      "loss": 0.4393,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.43908944725990295,
      "step": 445,
      "valid_targets_mean": 11036.9,
      "valid_targets_min": 2106
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 0.20232301950454712,
      "learning_rate": 1.580148432300241e-05,
      "loss": 0.4411,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.42553338408470154,
      "step": 450,
      "valid_targets_mean": 10535.1,
      "valid_targets_min": 2257
    },
    {
      "epoch": 4.333333333333333,
      "grad_norm": 0.19662712514400482,
      "learning_rate": 1.5338025221018668e-05,
      "loss": 0.445,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4404326379299164,
      "step": 455,
      "valid_targets_mean": 11668.5,
      "valid_targets_min": 1915
    },
    {
      "epoch": 4.380952380952381,
      "grad_norm": 0.1773589700460434,
      "learning_rate": 1.4877198727761748e-05,
      "loss": 0.4454,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45294705033302307,
      "step": 460,
      "valid_targets_mean": 11352.0,
      "valid_targets_min": 518
    },
    {
      "epoch": 4.428571428571429,
      "grad_norm": 0.21333052217960358,
      "learning_rate": 1.4419265071119038e-05,
      "loss": 0.4374,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.424248069524765,
      "step": 465,
      "valid_targets_mean": 9814.9,
      "valid_targets_min": 1652
    },
    {
      "epoch": 4.476190476190476,
      "grad_norm": 0.20808833837509155,
      "learning_rate": 1.3964482845398281e-05,
      "loss": 0.4501,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4347710907459259,
      "step": 470,
      "valid_targets_mean": 10299.0,
      "valid_targets_min": 1547
    },
    {
      "epoch": 4.523809523809524,
      "grad_norm": 0.20887579023838043,
      "learning_rate": 1.3513108865299907e-05,
      "loss": 0.4441,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4441516697406769,
      "step": 475,
      "valid_targets_mean": 11044.6,
      "valid_targets_min": 1307
    },
    {
      "epoch": 4.571428571428571,
      "grad_norm": 0.22309023141860962,
      "learning_rate": 1.3065398020894202e-05,
      "loss": 0.4433,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4627133309841156,
      "step": 480,
      "valid_targets_mean": 11029.4,
      "valid_targets_min": 1539
    },
    {
      "epoch": 4.619047619047619,
      "grad_norm": 0.20956307649612427,
      "learning_rate": 1.2621603133685343e-05,
      "loss": 0.4434,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44400516152381897,
      "step": 485,
      "valid_targets_mean": 9374.1,
      "valid_targets_min": 971
    },
    {
      "epoch": 4.666666666666667,
      "grad_norm": 0.19290389120578766,
      "learning_rate": 1.218197481384356e-05,
      "loss": 0.449,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.43824276328086853,
      "step": 490,
      "valid_targets_mean": 11172.0,
      "valid_targets_min": 1773
    },
    {
      "epoch": 4.714285714285714,
      "grad_norm": 0.20596766471862793,
      "learning_rate": 1.1746761318686044e-05,
      "loss": 0.4405,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4291389286518097,
      "step": 495,
      "valid_targets_mean": 10622.1,
      "valid_targets_min": 996
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 0.18378686904907227,
      "learning_rate": 1.1316208412486443e-05,
      "loss": 0.4486,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44633662700653076,
      "step": 500,
      "valid_targets_mean": 11159.3,
      "valid_targets_min": 1407
    },
    {
      "epoch": 4.809523809523809,
      "grad_norm": 0.1979859471321106,
      "learning_rate": 1.0890559227692265e-05,
      "loss": 0.4455,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4464368522167206,
      "step": 505,
      "valid_targets_mean": 9961.4,
      "valid_targets_min": 1277
    },
    {
      "epoch": 4.857142857142857,
      "grad_norm": 0.25758323073387146,
      "learning_rate": 1.0470054127628411e-05,
      "loss": 0.4414,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4472590386867523,
      "step": 510,
      "valid_targets_mean": 11638.4,
      "valid_targets_min": 1695
    },
    {
      "epoch": 4.904761904761905,
      "grad_norm": 0.2068321853876114,
      "learning_rate": 1.0054930570764427e-05,
      "loss": 0.4416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4409043788909912,
      "step": 515,
      "valid_targets_mean": 9860.2,
      "valid_targets_min": 1193
    },
    {
      "epoch": 4.9523809523809526,
      "grad_norm": 0.22515769302845,
      "learning_rate": 9.645422976622154e-06,
      "loss": 0.4358,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4464632570743561,
      "step": 520,
      "valid_targets_mean": 10226.5,
      "valid_targets_min": 346
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.22349441051483154,
      "learning_rate": 9.241762593399437e-06,
      "loss": 0.4431,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4462651014328003,
      "step": 525,
      "valid_targets_mean": 10547.0,
      "valid_targets_min": 267
    },
    {
      "epoch": 5.0476190476190474,
      "grad_norm": 0.19069649279117584,
      "learning_rate": 8.844177367384689e-06,
      "loss": 0.44,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.43878093361854553,
      "step": 530,
      "valid_targets_mean": 9826.1,
      "valid_targets_min": 1517
    },
    {
      "epoch": 5.095238095238095,
      "grad_norm": 0.176970973610878,
      "learning_rate": 8.452891814236037e-06,
      "loss": 0.4382,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4395877420902252,
      "step": 535,
      "valid_targets_mean": 11575.5,
      "valid_targets_min": 1765
    },
    {
      "epoch": 5.142857142857143,
      "grad_norm": 0.18036945164203644,
      "learning_rate": 8.068126892197728e-06,
      "loss": 0.4361,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.439093679189682,
      "step": 540,
      "valid_targets_mean": 10604.6,
      "valid_targets_min": 1157
    },
    {
      "epoch": 5.190476190476191,
      "grad_norm": 0.180902361869812,
      "learning_rate": 7.690099877325419e-06,
      "loss": 0.4504,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4543825685977936,
      "step": 545,
      "valid_targets_mean": 11103.2,
      "valid_targets_min": 1392
    },
    {
      "epoch": 5.238095238095238,
      "grad_norm": 0.1751468926668167,
      "learning_rate": 7.319024240790768e-06,
      "loss": 0.4427,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4390667676925659,
      "step": 550,
      "valid_targets_mean": 10293.1,
      "valid_targets_min": 1243
    },
    {
      "epoch": 5.285714285714286,
      "grad_norm": 0.173446387052536,
      "learning_rate": 6.955109528334667e-06,
      "loss": 0.4312,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4159672260284424,
      "step": 555,
      "valid_targets_mean": 9913.7,
      "valid_targets_min": 1938
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 0.18824054300785065,
      "learning_rate": 6.59856124193712e-06,
      "loss": 0.4383,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44193482398986816,
      "step": 560,
      "valid_targets_mean": 9919.7,
      "valid_targets_min": 2123
    },
    {
      "epoch": 5.380952380952381,
      "grad_norm": 0.1922815442085266,
      "learning_rate": 6.249580723770665e-06,
      "loss": 0.4398,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44894376397132874,
      "step": 565,
      "valid_targets_mean": 10213.7,
      "valid_targets_min": 1250
    },
    {
      "epoch": 5.428571428571429,
      "grad_norm": 0.17625091969966888,
      "learning_rate": 5.908365042502801e-06,
      "loss": 0.4401,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.42889514565467834,
      "step": 570,
      "valid_targets_mean": 11648.2,
      "valid_targets_min": 1535
    },
    {
      "epoch": 5.476190476190476,
      "grad_norm": 0.19493943452835083,
      "learning_rate": 5.5751068820116784e-06,
      "loss": 0.438,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4378260672092438,
      "step": 575,
      "valid_targets_mean": 11193.7,
      "valid_targets_min": 2288
    },
    {
      "epoch": 5.523809523809524,
      "grad_norm": 0.21737957000732422,
      "learning_rate": 5.24999443257785e-06,
      "loss": 0.4446,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.46308979392051697,
      "step": 580,
      "valid_targets_mean": 10749.8,
      "valid_targets_min": 1639
    },
    {
      "epoch": 5.571428571428571,
      "grad_norm": 0.2082742601633072,
      "learning_rate": 4.9332112846135664e-06,
      "loss": 0.4345,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4396940767765045,
      "step": 585,
      "valid_targets_mean": 9398.3,
      "valid_targets_min": 1775
    },
    {
      "epoch": 5.619047619047619,
      "grad_norm": 0.18518823385238647,
      "learning_rate": 4.624936324989602e-06,
      "loss": 0.4297,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.42118358612060547,
      "step": 590,
      "valid_targets_mean": 9676.2,
      "valid_targets_min": 1814
    },
    {
      "epoch": 5.666666666666667,
      "grad_norm": 0.18208129703998566,
      "learning_rate": 4.325343636018165e-06,
      "loss": 0.4359,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4350763261318207,
      "step": 595,
      "valid_targets_mean": 10548.8,
      "valid_targets_min": 309
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 0.17130884528160095,
      "learning_rate": 4.0346023971489215e-06,
      "loss": 0.4379,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4484126567840576,
      "step": 600,
      "valid_targets_mean": 10558.0,
      "valid_targets_min": 1478
    },
    {
      "epoch": 5.761904761904762,
      "grad_norm": 0.18925878405570984,
      "learning_rate": 3.752876789433677e-06,
      "loss": 0.4431,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45218753814697266,
      "step": 605,
      "valid_targets_mean": 10494.7,
      "valid_targets_min": 1629
    },
    {
      "epoch": 5.809523809523809,
      "grad_norm": 0.17503197491168976,
      "learning_rate": 3.480325902813624e-06,
      "loss": 0.4364,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44668635725975037,
      "step": 610,
      "valid_targets_mean": 11100.2,
      "valid_targets_min": 1652
    },
    {
      "epoch": 5.857142857142857,
      "grad_norm": 0.18204058706760406,
      "learning_rate": 3.2171036462815563e-06,
      "loss": 0.4416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4363596737384796,
      "step": 615,
      "valid_targets_mean": 10854.3,
      "valid_targets_min": 1515
    },
    {
      "epoch": 5.904761904761905,
      "grad_norm": 0.16673018038272858,
      "learning_rate": 2.9633586609697086e-06,
      "loss": 0.4387,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45320454239845276,
      "step": 620,
      "valid_targets_mean": 11283.6,
      "valid_targets_min": 1778
    },
    {
      "epoch": 5.9523809523809526,
      "grad_norm": 0.1611456274986267,
      "learning_rate": 2.7192342362124048e-06,
      "loss": 0.4358,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44078150391578674,
      "step": 625,
      "valid_targets_mean": 10528.4,
      "valid_targets_min": 995
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.17028920352458954,
      "learning_rate": 2.4848682286308346e-06,
      "loss": 0.4312,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.43743446469306946,
      "step": 630,
      "valid_targets_mean": 10760.8,
      "valid_targets_min": 1386
    },
    {
      "epoch": 6.0476190476190474,
      "grad_norm": 0.18112342059612274,
      "learning_rate": 2.260392984285633e-06,
      "loss": 0.4317,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4303453862667084,
      "step": 635,
      "valid_targets_mean": 9819.8,
      "valid_targets_min": 1228
    },
    {
      "epoch": 6.095238095238095,
      "grad_norm": 0.16256897151470184,
      "learning_rate": 2.0459352639413343e-06,
      "loss": 0.4369,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.42814159393310547,
      "step": 640,
      "valid_targets_mean": 10641.3,
      "valid_targets_min": 1060
    },
    {
      "epoch": 6.142857142857143,
      "grad_norm": 0.20285063982009888,
      "learning_rate": 1.841616171484797e-06,
      "loss": 0.4371,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4268749952316284,
      "step": 645,
      "valid_targets_mean": 9974.1,
      "valid_targets_min": 1245
    },
    {
      "epoch": 6.190476190476191,
      "grad_norm": 0.17999224364757538,
      "learning_rate": 1.6475510855380195e-06,
      "loss": 0.4383,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4317034184932709,
      "step": 650,
      "valid_targets_mean": 10107.1,
      "valid_targets_min": 1695
    },
    {
      "epoch": 6.238095238095238,
      "grad_norm": 0.1909116506576538,
      "learning_rate": 1.4638495943040854e-06,
      "loss": 0.4387,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4486290514469147,
      "step": 655,
      "valid_targets_mean": 10502.2,
      "valid_targets_min": 1610
    },
    {
      "epoch": 6.285714285714286,
      "grad_norm": 0.17877644300460815,
      "learning_rate": 1.2906154336828913e-06,
      "loss": 0.4325,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4392260015010834,
      "step": 660,
      "valid_targets_mean": 10547.3,
      "valid_targets_min": 975
    },
    {
      "epoch": 6.333333333333333,
      "grad_norm": 0.1902361363172531,
      "learning_rate": 1.1279464286916508e-06,
      "loss": 0.4324,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.427330881357193,
      "step": 665,
      "valid_targets_mean": 10113.3,
      "valid_targets_min": 1520
    },
    {
      "epoch": 6.380952380952381,
      "grad_norm": 0.16514748334884644,
      "learning_rate": 9.759344382233048e-07,
      "loss": 0.429,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.42980971932411194,
      "step": 670,
      "valid_targets_mean": 10894.8,
      "valid_targets_min": 2130
    },
    {
      "epoch": 6.428571428571429,
      "grad_norm": 0.17610912024974823,
      "learning_rate": 8.34665303173976e-07,
      "loss": 0.439,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.45283421874046326,
      "step": 675,
      "valid_targets_mean": 10147.8,
      "valid_targets_min": 1334
    },
    {
      "epoch": 6.476190476190476,
      "grad_norm": 0.18840855360031128,
      "learning_rate": 7.042187979687432e-07,
      "loss": 0.4418,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44663646817207336,
      "step": 680,
      "valid_targets_mean": 10371.0,
      "valid_targets_min": 267
    },
    {
      "epoch": 6.523809523809524,
      "grad_norm": 0.15983620285987854,
      "learning_rate": 5.846685855131929e-07,
      "loss": 0.4352,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4321114718914032,
      "step": 685,
      "valid_targets_mean": 11962.8,
      "valid_targets_min": 1591
    },
    {
      "epoch": 6.571428571428571,
      "grad_norm": 0.17415718734264374,
      "learning_rate": 4.760821755961065e-07,
      "loss": 0.4375,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.443795770406723,
      "step": 690,
      "valid_targets_mean": 9948.1,
      "valid_targets_min": 1135
    },
    {
      "epoch": 6.619047619047619,
      "grad_norm": 0.17314453423023224,
      "learning_rate": 3.7852088676678665e-07,
      "loss": 0.4339,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.42952966690063477,
      "step": 695,
      "valid_targets_mean": 10110.8,
      "valid_targets_min": 1481
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.1725008487701416,
      "learning_rate": 2.920398117086043e-07,
      "loss": 0.435,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.43847766518592834,
      "step": 700,
      "valid_targets_mean": 10822.0,
      "valid_targets_min": 1197
    },
    {
      "epoch": 6.714285714285714,
      "grad_norm": 0.17323048412799835,
      "learning_rate": 2.1668778612825347e-07,
      "loss": 0.4332,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.44469061493873596,
      "step": 705,
      "valid_targets_mean": 11074.7,
      "valid_targets_min": 1368
    },
    {
      "epoch": 6.761904761904762,
      "grad_norm": 0.1832982748746872,
      "learning_rate": 1.5250736117830455e-07,
      "loss": 0.4404,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4256347119808197,
      "step": 710,
      "valid_targets_mean": 9906.4,
      "valid_targets_min": 2598
    },
    {
      "epoch": 6.809523809523809,
      "grad_norm": 0.16123178601264954,
      "learning_rate": 9.953477942866052e-08,
      "loss": 0.4369,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.43369975686073303,
      "step": 715,
      "valid_targets_mean": 10846.9,
      "valid_targets_min": 1448
    },
    {
      "epoch": 6.857142857142857,
      "grad_norm": 0.1590285748243332,
      "learning_rate": 5.779995440044594e-08,
      "loss": 0.4353,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4261643588542938,
      "step": 720,
      "valid_targets_mean": 10662.7,
      "valid_targets_min": 1696
    },
    {
      "epoch": 6.904761904761905,
      "grad_norm": 0.17242933809757233,
      "learning_rate": 2.7326453673872653e-08,
      "loss": 0.4411,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4442724883556366,
      "step": 725,
      "valid_targets_mean": 11019.9,
      "valid_targets_min": 2596
    },
    {
      "epoch": 6.9523809523809526,
      "grad_norm": 0.17852503061294556,
      "learning_rate": 8.131485579692121e-09,
      "loss": 0.4432,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4505399763584137,
      "step": 730,
      "valid_targets_mean": 10760.5,
      "valid_targets_min": 346
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.17249806225299835,
      "learning_rate": 2.2588948167756586e-10,
      "loss": 0.4358,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4335353374481201,
      "step": 735,
      "valid_targets_mean": 10050.6,
      "valid_targets_min": 1517
    },
    {
      "epoch": 7.0,
      "step": 735,
      "total_flos": 4706601917218816.0,
      "train_loss": 0.0,
      "train_runtime": 3.6306,
      "train_samples_per_second": 19280.632,
      "train_steps_per_second": 202.447
    }
  ],
  "logging_steps": 5,
  "max_steps": 735,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 400,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4706601917218816.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}