{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 1632,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02457002457002457,
      "grad_norm": 3.5317310427201827,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.8461,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.2849353849887848,
      "step": 5,
      "valid_targets_mean": 1490.1,
      "valid_targets_min": 971
    },
    {
      "epoch": 0.04914004914004914,
      "grad_norm": 1.202219130090371,
      "learning_rate": 0.0001,
      "loss": 0.4291,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.2022167444229126,
      "step": 10,
      "valid_targets_mean": 1436.6,
      "valid_targets_min": 853
    },
    {
      "epoch": 0.07371007371007371,
      "grad_norm": 0.8787295845972538,
      "learning_rate": 9.999765825644824e-05,
      "loss": 0.335,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1640535295009613,
      "step": 15,
      "valid_targets_mean": 1385.0,
      "valid_targets_min": 568
    },
    {
      "epoch": 0.09828009828009827,
      "grad_norm": 0.7147045925033308,
      "learning_rate": 9.999063324514344e-05,
      "loss": 0.3171,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15234357118606567,
      "step": 20,
      "valid_targets_mean": 1367.7,
      "valid_targets_min": 574
    },
    {
      "epoch": 0.12285012285012285,
      "grad_norm": 0.6497958620035181,
      "learning_rate": 9.99789256241166e-05,
      "loss": 0.285,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1415839046239853,
      "step": 25,
      "valid_targets_mean": 1316.3,
      "valid_targets_min": 783
    },
    {
      "epoch": 0.14742014742014742,
      "grad_norm": 0.5863502908920704,
      "learning_rate": 9.996253649001759e-05,
      "loss": 0.2762,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13542860746383667,
      "step": 30,
      "valid_targets_mean": 1348.1,
      "valid_targets_min": 693
    },
    {
      "epoch": 0.171990171990172,
      "grad_norm": 0.626750978778753,
      "learning_rate": 9.994146737801235e-05,
      "loss": 0.2814,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17099091410636902,
      "step": 35,
      "valid_targets_mean": 1599.6,
      "valid_targets_min": 768
    },
    {
      "epoch": 0.19656019656019655,
      "grad_norm": 0.5722325932529337,
      "learning_rate": 9.991572026163916e-05,
      "loss": 0.2653,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12752854824066162,
      "step": 40,
      "valid_targets_mean": 1151.1,
      "valid_targets_min": 712
    },
    {
      "epoch": 0.22113022113022113,
      "grad_norm": 0.8638136318346279,
      "learning_rate": 9.988529755262379e-05,
      "loss": 0.2596,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13361497223377228,
      "step": 45,
      "valid_targets_mean": 1315.8,
      "valid_targets_min": 649
    },
    {
      "epoch": 0.2457002457002457,
      "grad_norm": 1.8587267978944855,
      "learning_rate": 9.985020210065353e-05,
      "loss": 0.2706,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12216434627771378,
      "step": 50,
      "valid_targets_mean": 1322.4,
      "valid_targets_min": 940
    },
    {
      "epoch": 0.2702702702702703,
      "grad_norm": 0.547941104608437,
      "learning_rate": 9.981043719311034e-05,
      "loss": 0.2678,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14429020881652832,
      "step": 55,
      "valid_targets_mean": 1408.8,
      "valid_targets_min": 897
    },
    {
      "epoch": 0.29484029484029484,
      "grad_norm": 0.5986557075278855,
      "learning_rate": 9.976600655476283e-05,
      "loss": 0.2557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14186960458755493,
      "step": 60,
      "valid_targets_mean": 1364.4,
      "valid_targets_min": 710
    },
    {
      "epoch": 0.3194103194103194,
      "grad_norm": 0.5897591965468908,
      "learning_rate": 9.971691434741742e-05,
      "loss": 0.2584,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11381076276302338,
      "step": 65,
      "valid_targets_mean": 1072.5,
      "valid_targets_min": 663
    },
    {
      "epoch": 0.343980343980344,
      "grad_norm": 0.5433018400955162,
      "learning_rate": 9.966316516952854e-05,
      "loss": 0.2649,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13451801240444183,
      "step": 70,
      "valid_targets_mean": 1287.7,
      "valid_targets_min": 637
    },
    {
      "epoch": 0.36855036855036855,
      "grad_norm": 0.47134946266510613,
      "learning_rate": 9.960476405576782e-05,
      "loss": 0.256,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12614522874355316,
      "step": 75,
      "valid_targets_mean": 1355.8,
      "valid_targets_min": 646
    },
    {
      "epoch": 0.3931203931203931,
      "grad_norm": 0.5009575392578494,
      "learning_rate": 9.95417164765525e-05,
      "loss": 0.2568,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11249350011348724,
      "step": 80,
      "valid_targets_mean": 1198.3,
      "valid_targets_min": 800
    },
    {
      "epoch": 0.4176904176904177,
      "grad_norm": 0.4964813124723332,
      "learning_rate": 9.947402833753307e-05,
      "loss": 0.2573,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13899488747119904,
      "step": 85,
      "valid_targets_mean": 1440.5,
      "valid_targets_min": 598
    },
    {
      "epoch": 0.44226044226044225,
      "grad_norm": 0.52664451404203,
      "learning_rate": 9.940170597904006e-05,
      "loss": 0.2607,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1289326548576355,
      "step": 90,
      "valid_targets_mean": 1178.6,
      "valid_targets_min": 721
    },
    {
      "epoch": 0.4668304668304668,
      "grad_norm": 0.47578073110132335,
      "learning_rate": 9.932475617549016e-05,
      "loss": 0.2513,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13900956511497498,
      "step": 95,
      "valid_targets_mean": 1329.9,
      "valid_targets_min": 619
    },
    {
      "epoch": 0.4914004914004914,
      "grad_norm": 0.4555196443562555,
      "learning_rate": 9.924318613475156e-05,
      "loss": 0.2424,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09756769239902496,
      "step": 100,
      "valid_targets_mean": 1299.4,
      "valid_targets_min": 618
    },
    {
      "epoch": 0.515970515970516,
      "grad_norm": 0.4403223130979152,
      "learning_rate": 9.915700349746898e-05,
      "loss": 0.253,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12914995849132538,
      "step": 105,
      "valid_targets_mean": 1420.3,
      "valid_targets_min": 861
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 0.45106359259219175,
      "learning_rate": 9.906621633634782e-05,
      "loss": 0.2423,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1240796148777008,
      "step": 110,
      "valid_targets_mean": 1370.2,
      "valid_targets_min": 783
    },
    {
      "epoch": 0.5651105651105651,
      "grad_norm": 0.48058617450569124,
      "learning_rate": 9.897083315539803e-05,
      "loss": 0.2469,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12699732184410095,
      "step": 115,
      "valid_targets_mean": 1312.5,
      "valid_targets_min": 809
    },
    {
      "epoch": 0.5896805896805897,
      "grad_norm": 0.4221194167388693,
      "learning_rate": 9.88708628891376e-05,
      "loss": 0.2463,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11394324898719788,
      "step": 120,
      "valid_targets_mean": 1403.4,
      "valid_targets_min": 631
    },
    {
      "epoch": 0.6142506142506142,
      "grad_norm": 0.4610537918913265,
      "learning_rate": 9.876631490175555e-05,
      "loss": 0.2502,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12918531894683838,
      "step": 125,
      "valid_targets_mean": 1384.2,
      "valid_targets_min": 643
    },
    {
      "epoch": 0.6388206388206388,
      "grad_norm": 0.42247998299685735,
      "learning_rate": 9.86571989862349e-05,
      "loss": 0.2502,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11708711087703705,
      "step": 130,
      "valid_targets_mean": 1212.7,
      "valid_targets_min": 577
    },
    {
      "epoch": 0.6633906633906634,
      "grad_norm": 0.42116965624763975,
      "learning_rate": 9.854352536343534e-05,
      "loss": 0.2503,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14355707168579102,
      "step": 135,
      "valid_targets_mean": 1577.8,
      "valid_targets_min": 753
    },
    {
      "epoch": 0.687960687960688,
      "grad_norm": 0.596658705056403,
      "learning_rate": 9.842530468113578e-05,
      "loss": 0.2456,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.113030344247818,
      "step": 140,
      "valid_targets_mean": 1120.2,
      "valid_targets_min": 631
    },
    {
      "epoch": 0.7125307125307125,
      "grad_norm": 0.44263538751073955,
      "learning_rate": 9.830254801303702e-05,
      "loss": 0.2416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11774349212646484,
      "step": 145,
      "valid_targets_mean": 1205.3,
      "valid_targets_min": 833
    },
    {
      "epoch": 0.7371007371007371,
      "grad_norm": 0.40787430941229114,
      "learning_rate": 9.817526685772452e-05,
      "loss": 0.2437,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12346543371677399,
      "step": 150,
      "valid_targets_mean": 1249.8,
      "valid_targets_min": 656
    },
    {
      "epoch": 0.7616707616707616,
      "grad_norm": 0.4196828968071708,
      "learning_rate": 9.804347313759126e-05,
      "loss": 0.239,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11186127364635468,
      "step": 155,
      "valid_targets_mean": 1366.1,
      "valid_targets_min": 828
    },
    {
      "epoch": 0.7862407862407862,
      "grad_norm": 0.38233087493681384,
      "learning_rate": 9.790717919772102e-05,
      "loss": 0.23,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11590471118688583,
      "step": 160,
      "valid_targets_mean": 1399.3,
      "valid_targets_min": 848
    },
    {
      "epoch": 0.8108108108108109,
      "grad_norm": 0.3840556079978897,
      "learning_rate": 9.776639780473198e-05,
      "loss": 0.2354,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11734741926193237,
      "step": 165,
      "valid_targets_mean": 1268.5,
      "valid_targets_min": 559
    },
    {
      "epoch": 0.8353808353808354,
      "grad_norm": 0.4109296293583164,
      "learning_rate": 9.762114214558092e-05,
      "loss": 0.2412,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11579617857933044,
      "step": 170,
      "valid_targets_mean": 1429.6,
      "valid_targets_min": 613
    },
    {
      "epoch": 0.85995085995086,
      "grad_norm": 0.37286715566379114,
      "learning_rate": 9.747142582632795e-05,
      "loss": 0.2466,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11519800126552582,
      "step": 175,
      "valid_targets_mean": 1256.2,
      "valid_targets_min": 592
    },
    {
      "epoch": 0.8845208845208845,
      "grad_norm": 0.3611949235146365,
      "learning_rate": 9.731726287086211e-05,
      "loss": 0.2439,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11991137266159058,
      "step": 180,
      "valid_targets_mean": 1582.8,
      "valid_targets_min": 696
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.39325905281989476,
      "learning_rate": 9.715866771958766e-05,
      "loss": 0.2381,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11788970232009888,
      "step": 185,
      "valid_targets_mean": 1309.1,
      "valid_targets_min": 601
    },
    {
      "epoch": 0.9336609336609336,
      "grad_norm": 0.4568668063950179,
      "learning_rate": 9.699565522807151e-05,
      "loss": 0.2418,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1191166564822197,
      "step": 190,
      "valid_targets_mean": 1381.5,
      "valid_targets_min": 855
    },
    {
      "epoch": 0.9582309582309583,
      "grad_norm": 0.3969433794676836,
      "learning_rate": 9.682824066565168e-05,
      "loss": 0.2312,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1245565116405487,
      "step": 195,
      "valid_targets_mean": 1291.5,
      "valid_targets_min": 639
    },
    {
      "epoch": 0.9828009828009828,
      "grad_norm": 0.40452504567463415,
      "learning_rate": 9.665643971400709e-05,
      "loss": 0.2278,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12400303781032562,
      "step": 200,
      "valid_targets_mean": 1270.6,
      "valid_targets_min": 668
    },
    {
      "epoch": 1.0049140049140048,
      "grad_norm": 0.4019161541536171,
      "learning_rate": 9.648026846568853e-05,
      "loss": 0.2301,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10288889706134796,
      "step": 205,
      "valid_targets_mean": 1385.6,
      "valid_targets_min": 673
    },
    {
      "epoch": 1.0294840294840295,
      "grad_norm": 0.4713964035318676,
      "learning_rate": 9.629974342261142e-05,
      "loss": 0.188,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08235287666320801,
      "step": 210,
      "valid_targets_mean": 1291.2,
      "valid_targets_min": 658
    },
    {
      "epoch": 1.054054054054054,
      "grad_norm": 0.3728121128850294,
      "learning_rate": 9.611488149450995e-05,
      "loss": 0.179,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07877619564533234,
      "step": 215,
      "valid_targets_mean": 1312.9,
      "valid_targets_min": 717
    },
    {
      "epoch": 1.0786240786240786,
      "grad_norm": 0.39178811104454536,
      "learning_rate": 9.592569999735325e-05,
      "loss": 0.1728,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09154706448316574,
      "step": 220,
      "valid_targets_mean": 1480.9,
      "valid_targets_min": 746
    },
    {
      "epoch": 1.1031941031941033,
      "grad_norm": 0.3386346510433248,
      "learning_rate": 9.57322166517234e-05,
      "loss": 0.1787,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09614543616771698,
      "step": 225,
      "valid_targets_mean": 1499.4,
      "valid_targets_min": 873
    },
    {
      "epoch": 1.1277641277641277,
      "grad_norm": 0.39955268162367374,
      "learning_rate": 9.553444958115545e-05,
      "loss": 0.1794,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0914674922823906,
      "step": 230,
      "valid_targets_mean": 1553.2,
      "valid_targets_min": 975
    },
    {
      "epoch": 1.1523341523341524,
      "grad_norm": 0.37356699705679897,
      "learning_rate": 9.53324173104399e-05,
      "loss": 0.1758,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09144826233386993,
      "step": 235,
      "valid_targets_mean": 1287.4,
      "valid_targets_min": 682
    },
    {
      "epoch": 1.1769041769041768,
      "grad_norm": 0.3786464677111562,
      "learning_rate": 9.512613876388742e-05,
      "loss": 0.1767,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0818377286195755,
      "step": 240,
      "valid_targets_mean": 1352.4,
      "valid_targets_min": 800
    },
    {
      "epoch": 1.2014742014742015,
      "grad_norm": 0.36472496386857967,
      "learning_rate": 9.491563326355628e-05,
      "loss": 0.1798,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08443672955036163,
      "step": 245,
      "valid_targets_mean": 1144.6,
      "valid_targets_min": 706
    },
    {
      "epoch": 1.2260442260442261,
      "grad_norm": 0.4060906953722515,
      "learning_rate": 9.47009205274424e-05,
      "loss": 0.1826,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0913555771112442,
      "step": 250,
      "valid_targets_mean": 1341.1,
      "valid_targets_min": 729
    },
    {
      "epoch": 1.2506142506142506,
      "grad_norm": 0.38184644361730635,
      "learning_rate": 9.448202066763237e-05,
      "loss": 0.1814,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08872812986373901,
      "step": 255,
      "valid_targets_mean": 1251.3,
      "valid_targets_min": 761
    },
    {
      "epoch": 1.2751842751842752,
      "grad_norm": 0.35294094367142664,
      "learning_rate": 9.425895418841961e-05,
      "loss": 0.1799,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07888481020927429,
      "step": 260,
      "valid_targets_mean": 1164.6,
      "valid_targets_min": 663
    },
    {
      "epoch": 1.2997542997542997,
      "grad_norm": 0.36613096681483226,
      "learning_rate": 9.403174198438372e-05,
      "loss": 0.1765,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08659601211547852,
      "step": 265,
      "valid_targets_mean": 1139.0,
      "valid_targets_min": 614
    },
    {
      "epoch": 1.3243243243243243,
      "grad_norm": 0.38382168587316545,
      "learning_rate": 9.380040533843319e-05,
      "loss": 0.1791,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09439398348331451,
      "step": 270,
      "valid_targets_mean": 1348.4,
      "valid_targets_min": 854
    },
    {
      "epoch": 1.348894348894349,
      "grad_norm": 0.3642425830054166,
      "learning_rate": 9.356496591981204e-05,
      "loss": 0.1775,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08763645589351654,
      "step": 275,
      "valid_targets_mean": 1346.3,
      "valid_targets_min": 650
    },
    {
      "epoch": 1.3734643734643734,
      "grad_norm": 0.3832204309855803,
      "learning_rate": 9.332544578206985e-05,
      "loss": 0.1771,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08698828518390656,
      "step": 280,
      "valid_targets_mean": 1353.8,
      "valid_targets_min": 623
    },
    {
      "epoch": 1.398034398034398,
      "grad_norm": 0.3561629333331951,
      "learning_rate": 9.308186736099614e-05,
      "loss": 0.1838,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09670018404722214,
      "step": 285,
      "valid_targets_mean": 1320.3,
      "valid_targets_min": 781
    },
    {
      "epoch": 1.4226044226044225,
      "grad_norm": 0.3765510806090441,
      "learning_rate": 9.28342534725188e-05,
      "loss": 0.1823,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08860672265291214,
      "step": 290,
      "valid_targets_mean": 1303.7,
      "valid_targets_min": 582
    },
    {
      "epoch": 1.4471744471744472,
      "grad_norm": 0.33661939310832245,
      "learning_rate": 9.258262731056688e-05,
      "loss": 0.1769,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08029186725616455,
      "step": 295,
      "valid_targets_mean": 1230.8,
      "valid_targets_min": 688
    },
    {
      "epoch": 1.4717444717444716,
      "grad_norm": 0.3643823226415973,
      "learning_rate": 9.23270124448981e-05,
      "loss": 0.1814,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0855192244052887,
      "step": 300,
      "valid_targets_mean": 1241.6,
      "valid_targets_min": 684
    },
    {
      "epoch": 1.4963144963144963,
      "grad_norm": 0.36001426651444374,
      "learning_rate": 9.206743281889097e-05,
      "loss": 0.1741,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08223570883274078,
      "step": 305,
      "valid_targets_mean": 1359.8,
      "valid_targets_min": 633
    },
    {
      "epoch": 1.520884520884521,
      "grad_norm": 0.3622448158704566,
      "learning_rate": 9.18039127473021e-05,
      "loss": 0.1825,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08101462572813034,
      "step": 310,
      "valid_targets_mean": 1251.8,
      "valid_targets_min": 502
    },
    {
      "epoch": 1.5454545454545454,
      "grad_norm": 0.3673440986818992,
      "learning_rate": 9.153647691398866e-05,
      "loss": 0.1824,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08977461606264114,
      "step": 315,
      "valid_targets_mean": 1314.8,
      "valid_targets_min": 650
    },
    {
      "epoch": 1.57002457002457,
      "grad_norm": 0.3712553762342498,
      "learning_rate": 9.126515036959613e-05,
      "loss": 0.1834,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0836411640048027,
      "step": 320,
      "valid_targets_mean": 1373.9,
      "valid_targets_min": 741
    },
    {
      "epoch": 1.5945945945945947,
      "grad_norm": 0.3641331948462104,
      "learning_rate": 9.098995852921197e-05,
      "loss": 0.1788,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08982405811548233,
      "step": 325,
      "valid_targets_mean": 1291.5,
      "valid_targets_min": 654
    },
    {
      "epoch": 1.6191646191646192,
      "grad_norm": 0.34273897831427635,
      "learning_rate": 9.07109271699849e-05,
      "loss": 0.1883,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07989470660686493,
      "step": 330,
      "valid_targets_mean": 1270.9,
      "valid_targets_min": 590
    },
    {
      "epoch": 1.6437346437346436,
      "grad_norm": 0.3849520242237801,
      "learning_rate": 9.042808242871035e-05,
      "loss": 0.1865,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09401971101760864,
      "step": 335,
      "valid_targets_mean": 1381.6,
      "valid_targets_min": 765
    },
    {
      "epoch": 1.6683046683046683,
      "grad_norm": 0.36040402759215207,
      "learning_rate": 9.014145079938228e-05,
      "loss": 0.1838,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08834733068943024,
      "step": 340,
      "valid_targets_mean": 1301.9,
      "valid_targets_min": 724
    },
    {
      "epoch": 1.692874692874693,
      "grad_norm": 0.35689718091019035,
      "learning_rate": 8.985105913071148e-05,
      "loss": 0.1787,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10028181225061417,
      "step": 345,
      "valid_targets_mean": 1398.2,
      "valid_targets_min": 725
    },
    {
      "epoch": 1.7174447174447174,
      "grad_norm": 0.3615825498096839,
      "learning_rate": 8.955693462361065e-05,
      "loss": 0.179,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09255008399486542,
      "step": 350,
      "valid_targets_mean": 1295.4,
      "valid_targets_min": 803
    },
    {
      "epoch": 1.742014742014742,
      "grad_norm": 0.35872374357460324,
      "learning_rate": 8.925910482864652e-05,
      "loss": 0.1877,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0775061845779419,
      "step": 355,
      "valid_targets_mean": 1276.3,
      "valid_targets_min": 631
    },
    {
      "epoch": 1.7665847665847667,
      "grad_norm": 0.35482753893532454,
      "learning_rate": 8.895759764345914e-05,
      "loss": 0.1828,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09774541854858398,
      "step": 360,
      "valid_targets_mean": 1289.6,
      "valid_targets_min": 622
    },
    {
      "epoch": 1.7911547911547911,
      "grad_norm": 0.3785565441243277,
      "learning_rate": 8.865244131014883e-05,
      "loss": 0.182,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0870940238237381,
      "step": 365,
      "valid_targets_mean": 1332.1,
      "valid_targets_min": 626
    },
    {
      "epoch": 1.8157248157248156,
      "grad_norm": 0.3552269775568485,
      "learning_rate": 8.834366441263056e-05,
      "loss": 0.1781,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08362573385238647,
      "step": 370,
      "valid_targets_mean": 1125.2,
      "valid_targets_min": 521
    },
    {
      "epoch": 1.8402948402948403,
      "grad_norm": 0.37478075988467835,
      "learning_rate": 8.803129587395673e-05,
      "loss": 0.181,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08486934006214142,
      "step": 375,
      "valid_targets_mean": 1216.6,
      "valid_targets_min": 486
    },
    {
      "epoch": 1.864864864864865,
      "grad_norm": 0.3769227355201233,
      "learning_rate": 8.771536495360776e-05,
      "loss": 0.1777,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09369044005870819,
      "step": 380,
      "valid_targets_mean": 1202.1,
      "valid_targets_min": 671
    },
    {
      "epoch": 1.8894348894348894,
      "grad_norm": 0.3428843010485239,
      "learning_rate": 8.739590124475148e-05,
      "loss": 0.1808,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08531995117664337,
      "step": 385,
      "valid_targets_mean": 1225.2,
      "valid_targets_min": 589
    },
    {
      "epoch": 1.914004914004914,
      "grad_norm": 0.3502029019390788,
      "learning_rate": 8.707293467147109e-05,
      "loss": 0.1809,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09207258373498917,
      "step": 390,
      "valid_targets_mean": 1262.0,
      "valid_targets_min": 909
    },
    {
      "epoch": 1.9385749385749387,
      "grad_norm": 0.33525127668618215,
      "learning_rate": 8.674649548596221e-05,
      "loss": 0.1778,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08499929308891296,
      "step": 395,
      "valid_targets_mean": 1225.2,
      "valid_targets_min": 527
    },
    {
      "epoch": 1.9631449631449631,
      "grad_norm": 0.3278244741049693,
      "learning_rate": 8.641661426569916e-05,
      "loss": 0.1777,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.080820232629776,
      "step": 400,
      "valid_targets_mean": 1319.4,
      "valid_targets_min": 618
    },
    {
      "epoch": 1.9877149877149876,
      "grad_norm": 0.3479090084518426,
      "learning_rate": 8.608332191057076e-05,
      "loss": 0.1758,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08339822292327881,
      "step": 405,
      "valid_targets_mean": 1284.8,
      "valid_targets_min": 820
    },
    {
      "epoch": 2.0098280098280097,
      "grad_norm": 0.33772815977939674,
      "learning_rate": 8.57466496399859e-05,
      "loss": 0.1508,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05549818277359009,
      "step": 410,
      "valid_targets_mean": 1236.8,
      "valid_targets_min": 698
    },
    {
      "epoch": 2.0343980343980346,
      "grad_norm": 0.35332743790390486,
      "learning_rate": 8.54066289899494e-05,
      "loss": 0.12,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05471636354923248,
      "step": 415,
      "valid_targets_mean": 1206.4,
      "valid_targets_min": 796
    },
    {
      "epoch": 2.058968058968059,
      "grad_norm": 0.34889197536174715,
      "learning_rate": 8.506329181010781e-05,
      "loss": 0.1139,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.052125804126262665,
      "step": 420,
      "valid_targets_mean": 1230.4,
      "valid_targets_min": 486
    },
    {
      "epoch": 2.0835380835380835,
      "grad_norm": 0.342810837053402,
      "learning_rate": 8.471667026076621e-05,
      "loss": 0.1151,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05554091930389404,
      "step": 425,
      "valid_targets_mean": 1376.2,
      "valid_targets_min": 683
    },
    {
      "epoch": 2.108108108108108,
      "grad_norm": 0.3599219497453591,
      "learning_rate": 8.436679680987571e-05,
      "loss": 0.1148,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.056023404002189636,
      "step": 430,
      "valid_targets_mean": 1327.0,
      "valid_targets_min": 755
    },
    {
      "epoch": 2.1326781326781328,
      "grad_norm": 0.37719592865909146,
      "learning_rate": 8.401370422999224e-05,
      "loss": 0.1172,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06519174575805664,
      "step": 435,
      "valid_targets_mean": 1529.8,
      "valid_targets_min": 863
    },
    {
      "epoch": 2.157248157248157,
      "grad_norm": 0.3457538844180661,
      "learning_rate": 8.365742559520669e-05,
      "loss": 0.1169,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0638711005449295,
      "step": 440,
      "valid_targets_mean": 1522.7,
      "valid_targets_min": 881
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.3756160940209798,
      "learning_rate": 8.329799427804683e-05,
      "loss": 0.1173,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05762581154704094,
      "step": 445,
      "valid_targets_mean": 1156.8,
      "valid_targets_min": 521
    },
    {
      "epoch": 2.2063882063882065,
      "grad_norm": 0.35582823840706373,
      "learning_rate": 8.293544394635149e-05,
      "loss": 0.1159,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0591716468334198,
      "step": 450,
      "valid_targets_mean": 1395.5,
      "valid_targets_min": 575
    },
    {
      "epoch": 2.230958230958231,
      "grad_norm": 0.3474647468975514,
      "learning_rate": 8.256980856011672e-05,
      "loss": 0.1164,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05874478816986084,
      "step": 455,
      "valid_targets_mean": 1361.8,
      "valid_targets_min": 754
    },
    {
      "epoch": 2.2555282555282554,
      "grad_norm": 0.33681306262455324,
      "learning_rate": 8.22011223683148e-05,
      "loss": 0.114,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05156775563955307,
      "step": 460,
      "valid_targets_mean": 1222.9,
      "valid_targets_min": 658
    },
    {
      "epoch": 2.2800982800982803,
      "grad_norm": 0.3470018928684881,
      "learning_rate": 8.182941990568626e-05,
      "loss": 0.1144,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05891558527946472,
      "step": 465,
      "valid_targets_mean": 1401.7,
      "valid_targets_min": 681
    },
    {
      "epoch": 2.3046683046683047,
      "grad_norm": 0.3383126654609071,
      "learning_rate": 8.145473598950489e-05,
      "loss": 0.12,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05846539884805679,
      "step": 470,
      "valid_targets_mean": 1435.5,
      "valid_targets_min": 686
    },
    {
      "epoch": 2.329238329238329,
      "grad_norm": 0.3743955050765521,
      "learning_rate": 8.107710571631648e-05,
      "loss": 0.1213,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0630626231431961,
      "step": 475,
      "valid_targets_mean": 1401.7,
      "valid_targets_min": 761
    },
    {
      "epoch": 2.3538083538083536,
      "grad_norm": 0.3479810802972055,
      "learning_rate": 8.06965644586513e-05,
      "loss": 0.1186,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05939309298992157,
      "step": 480,
      "valid_targets_mean": 1252.8,
      "valid_targets_min": 600
    },
    {
      "epoch": 2.3783783783783785,
      "grad_norm": 0.34502770257222604,
      "learning_rate": 8.031314786171083e-05,
      "loss": 0.1205,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05798602104187012,
      "step": 485,
      "valid_targets_mean": 1292.5,
      "valid_targets_min": 599
    },
    {
      "epoch": 2.402948402948403,
      "grad_norm": 0.361312416737643,
      "learning_rate": 7.99268918400288e-05,
      "loss": 0.1183,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05880609527230263,
      "step": 490,
      "valid_targets_mean": 1293.4,
      "valid_targets_min": 643
    },
    {
      "epoch": 2.4275184275184274,
      "grad_norm": 0.36520874389005514,
      "learning_rate": 7.953783257410713e-05,
      "loss": 0.1178,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06413908302783966,
      "step": 495,
      "valid_targets_mean": 1497.6,
      "valid_targets_min": 739
    },
    {
      "epoch": 2.4520884520884523,
      "grad_norm": 0.33310913231009204,
      "learning_rate": 7.914600650702691e-05,
      "loss": 0.1209,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0641617551445961,
      "step": 500,
      "valid_targets_mean": 1445.1,
      "valid_targets_min": 775
    },
    {
      "epoch": 2.4766584766584767,
      "grad_norm": 0.37018880178023517,
      "learning_rate": 7.875145034103479e-05,
      "loss": 0.1187,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05668133497238159,
      "step": 505,
      "valid_targets_mean": 1145.2,
      "valid_targets_min": 652
    },
    {
      "epoch": 2.501228501228501,
      "grad_norm": 0.3699366956357319,
      "learning_rate": 7.835420103410504e-05,
      "loss": 0.12,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05439180135726929,
      "step": 510,
      "valid_targets_mean": 1140.4,
      "valid_targets_min": 690
    },
    {
      "epoch": 2.5257985257985256,
      "grad_norm": 0.3493711971707187,
      "learning_rate": 7.795429579647781e-05,
      "loss": 0.1198,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05129605531692505,
      "step": 515,
      "valid_targets_mean": 1130.1,
      "valid_targets_min": 510
    },
    {
      "epoch": 2.5503685503685505,
      "grad_norm": 0.34249507252623157,
      "learning_rate": 7.755177208717356e-05,
      "loss": 0.1209,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06780637800693512,
      "step": 520,
      "valid_targets_mean": 1362.8,
      "valid_targets_min": 785
    },
    {
      "epoch": 2.574938574938575,
      "grad_norm": 0.3570510910221036,
      "learning_rate": 7.71466676104843e-05,
      "loss": 0.1183,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.050399065017700195,
      "step": 525,
      "valid_targets_mean": 1082.3,
      "valid_targets_min": 501
    },
    {
      "epoch": 2.5995085995085994,
      "grad_norm": 0.3364722657405281,
      "learning_rate": 7.673902031244189e-05,
      "loss": 0.1203,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.062017861753702164,
      "step": 530,
      "valid_targets_mean": 1341.2,
      "valid_targets_min": 662
    },
    {
      "epoch": 2.6240786240786242,
      "grad_norm": 0.36931412486889464,
      "learning_rate": 7.632886837726359e-05,
      "loss": 0.1205,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06189979240298271,
      "step": 535,
      "valid_targets_mean": 1222.2,
      "valid_targets_min": 679
    },
    {
      "epoch": 2.6486486486486487,
      "grad_norm": 0.338457123494396,
      "learning_rate": 7.591625022377537e-05,
      "loss": 0.1206,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05179202929139137,
      "step": 540,
      "valid_targets_mean": 1184.9,
      "valid_targets_min": 492
    },
    {
      "epoch": 2.673218673218673,
      "grad_norm": 0.3244686185474909,
      "learning_rate": 7.550120450181324e-05,
      "loss": 0.1184,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06525638699531555,
      "step": 545,
      "valid_targets_mean": 1495.0,
      "valid_targets_min": 754
    },
    {
      "epoch": 2.697788697788698,
      "grad_norm": 0.335778159582729,
      "learning_rate": 7.508377008860294e-05,
      "loss": 0.1187,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.048954833298921585,
      "step": 550,
      "valid_targets_mean": 1146.8,
      "valid_targets_min": 629
    },
    {
      "epoch": 2.7223587223587224,
      "grad_norm": 0.3423155775776973,
      "learning_rate": 7.466398608511826e-05,
      "loss": 0.122,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0645413026213646,
      "step": 555,
      "valid_targets_mean": 1507.3,
      "valid_targets_min": 708
    },
    {
      "epoch": 2.746928746928747,
      "grad_norm": 0.3486117813784432,
      "learning_rate": 7.424189181241856e-05,
      "loss": 0.1192,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05744887888431549,
      "step": 560,
      "valid_targets_mean": 1291.0,
      "valid_targets_min": 654
    },
    {
      "epoch": 2.7714987714987718,
      "grad_norm": 0.34392630396696766,
      "learning_rate": 7.381752680796547e-05,
      "loss": 0.1174,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05362129211425781,
      "step": 565,
      "valid_targets_mean": 1325.0,
      "valid_targets_min": 679
    },
    {
      "epoch": 2.796068796068796,
      "grad_norm": 0.33308716912129516,
      "learning_rate": 7.339093082191953e-05,
      "loss": 0.1163,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05184609442949295,
      "step": 570,
      "valid_targets_mean": 1360.9,
      "valid_targets_min": 585
    },
    {
      "epoch": 2.8206388206388207,
      "grad_norm": 0.3587925120059478,
      "learning_rate": 7.29621438134167e-05,
      "loss": 0.1229,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0640011578798294,
      "step": 575,
      "valid_targets_mean": 1299.4,
      "valid_targets_min": 735
    },
    {
      "epoch": 2.845208845208845,
      "grad_norm": 0.3411894303363057,
      "learning_rate": 7.253120594682547e-05,
      "loss": 0.1235,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.058491215109825134,
      "step": 580,
      "valid_targets_mean": 1390.1,
      "valid_targets_min": 816
    },
    {
      "epoch": 2.8697788697788695,
      "grad_norm": 0.32947595924960127,
      "learning_rate": 7.209815758798464e-05,
      "loss": 0.1219,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.057379547506570816,
      "step": 585,
      "valid_targets_mean": 1288.5,
      "valid_targets_min": 631
    },
    {
      "epoch": 2.8943488943488944,
      "grad_norm": 0.4917965558943038,
      "learning_rate": 7.166303930042233e-05,
      "loss": 0.1191,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0557168647646904,
      "step": 590,
      "valid_targets_mean": 1338.1,
      "valid_targets_min": 641
    },
    {
      "epoch": 2.918918918918919,
      "grad_norm": 0.34569685035267406,
      "learning_rate": 7.122589184155626e-05,
      "loss": 0.121,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06044552102684975,
      "step": 595,
      "valid_targets_mean": 1258.4,
      "valid_targets_min": 670
    },
    {
      "epoch": 2.9434889434889433,
      "grad_norm": 0.314176762801982,
      "learning_rate": 7.078675615887618e-05,
      "loss": 0.1211,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05718706548213959,
      "step": 600,
      "valid_targets_mean": 1445.6,
      "valid_targets_min": 538
    },
    {
      "epoch": 2.968058968058968,
      "grad_norm": 0.3242469563892565,
      "learning_rate": 7.034567338610819e-05,
      "loss": 0.1215,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06585130095481873,
      "step": 605,
      "valid_targets_mean": 1457.1,
      "valid_targets_min": 772
    },
    {
      "epoch": 2.9926289926289926,
      "grad_norm": 0.34892426388331144,
      "learning_rate": 6.990268483936189e-05,
      "loss": 0.1215,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06495223939418793,
      "step": 610,
      "valid_targets_mean": 1545.8,
      "valid_targets_min": 729
    },
    {
      "epoch": 3.0147420147420148,
      "grad_norm": 0.2977325071801822,
      "learning_rate": 6.945783201326015e-05,
      "loss": 0.0882,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.037302177399396896,
      "step": 615,
      "valid_targets_mean": 1406.4,
      "valid_targets_min": 681
    },
    {
      "epoch": 3.039312039312039,
      "grad_norm": 0.3861565704828333,
      "learning_rate": 6.901115657705246e-05,
      "loss": 0.0709,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.041658252477645874,
      "step": 620,
      "valid_targets_mean": 1367.2,
      "valid_targets_min": 673
    },
    {
      "epoch": 3.063882063882064,
      "grad_norm": 0.33429050333383475,
      "learning_rate": 6.856270037071176e-05,
      "loss": 0.0714,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.036533765494823456,
      "step": 625,
      "valid_targets_mean": 1446.1,
      "valid_targets_min": 745
    },
    {
      "epoch": 3.0884520884520885,
      "grad_norm": 0.406298735051371,
      "learning_rate": 6.811250540101517e-05,
      "loss": 0.0691,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03498723730444908,
      "step": 630,
      "valid_targets_mean": 1332.6,
      "valid_targets_min": 880
    },
    {
      "epoch": 3.113022113022113,
      "grad_norm": 0.30666504482556517,
      "learning_rate": 6.766061383760943e-05,
      "loss": 0.0691,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.030431220307946205,
      "step": 635,
      "valid_targets_mean": 1329.4,
      "valid_targets_min": 847
    },
    {
      "epoch": 3.1375921375921374,
      "grad_norm": 0.3347413979325829,
      "learning_rate": 6.72070680090607e-05,
      "loss": 0.0694,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.037094853818416595,
      "step": 640,
      "valid_targets_mean": 1259.8,
      "valid_targets_min": 690
    },
    {
      "epoch": 3.1621621621621623,
      "grad_norm": 0.35070823982092636,
      "learning_rate": 6.675191039888978e-05,
      "loss": 0.0735,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03559793531894684,
      "step": 645,
      "valid_targets_mean": 1238.9,
      "valid_targets_min": 683
    },
    {
      "epoch": 3.1867321867321867,
      "grad_norm": 0.3480670160346414,
      "learning_rate": 6.629518364159259e-05,
      "loss": 0.0731,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.034793753176927567,
      "step": 650,
      "valid_targets_mean": 1273.3,
      "valid_targets_min": 740
    },
    {
      "epoch": 3.211302211302211,
      "grad_norm": 0.3288719326795959,
      "learning_rate": 6.583693051864668e-05,
      "loss": 0.0743,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0357627272605896,
      "step": 655,
      "valid_targets_mean": 1355.0,
      "valid_targets_min": 673
    },
    {
      "epoch": 3.235872235872236,
      "grad_norm": 0.3325184929663535,
      "learning_rate": 6.537719395450391e-05,
      "loss": 0.0735,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.035944681614637375,
      "step": 660,
      "valid_targets_mean": 1264.7,
      "valid_targets_min": 586
    },
    {
      "epoch": 3.2604422604422605,
      "grad_norm": 0.35118501191130574,
      "learning_rate": 6.491601701256966e-05,
      "loss": 0.0725,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0333075150847435,
      "step": 665,
      "valid_targets_mean": 1220.9,
      "valid_targets_min": 632
    },
    {
      "epoch": 3.285012285012285,
      "grad_norm": 0.3516843877270984,
      "learning_rate": 6.44534428911691e-05,
      "loss": 0.0714,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.037736207246780396,
      "step": 670,
      "valid_targets_mean": 1361.2,
      "valid_targets_min": 521
    },
    {
      "epoch": 3.30958230958231,
      "grad_norm": 0.32212096835158077,
      "learning_rate": 6.398951491950089e-05,
      "loss": 0.0747,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03314409777522087,
      "step": 675,
      "valid_targets_mean": 1343.1,
      "valid_targets_min": 787
    },
    {
      "epoch": 3.3341523341523343,
      "grad_norm": 0.36619223426782116,
      "learning_rate": 6.352427655357848e-05,
      "loss": 0.0729,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03750663995742798,
      "step": 680,
      "valid_targets_mean": 1133.4,
      "valid_targets_min": 692
    },
    {
      "epoch": 3.3587223587223587,
      "grad_norm": 0.3712584851066574,
      "learning_rate": 6.30577713721596e-05,
      "loss": 0.0754,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0373421236872673,
      "step": 685,
      "valid_targets_mean": 1293.4,
      "valid_targets_min": 723
    },
    {
      "epoch": 3.383292383292383,
      "grad_norm": 0.3413543433956999,
      "learning_rate": 6.259004307266426e-05,
      "loss": 0.0744,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03667682409286499,
      "step": 690,
      "valid_targets_mean": 1444.6,
      "valid_targets_min": 988
    },
    {
      "epoch": 3.407862407862408,
      "grad_norm": 0.34966421597251157,
      "learning_rate": 6.212113546708165e-05,
      "loss": 0.0733,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03585509583353996,
      "step": 695,
      "valid_targets_mean": 1392.1,
      "valid_targets_min": 649
    },
    {
      "epoch": 3.4324324324324325,
      "grad_norm": 0.33436130188939295,
      "learning_rate": 6.165109247786624e-05,
      "loss": 0.0752,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03639696165919304,
      "step": 700,
      "valid_targets_mean": 1271.6,
      "valid_targets_min": 613
    },
    {
      "epoch": 3.457002457002457,
      "grad_norm": 0.3370440922395482,
      "learning_rate": 6.117995813382357e-05,
      "loss": 0.0733,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03243023902177811,
      "step": 705,
      "valid_targets_mean": 1190.7,
      "valid_targets_min": 668
    },
    {
      "epoch": 3.4815724815724813,
      "grad_norm": 0.34572473082483673,
      "learning_rate": 6.070777656598615e-05,
      "loss": 0.0734,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03953058272600174,
      "step": 710,
      "valid_targets_mean": 1412.8,
      "valid_targets_min": 608
    },
    {
      "epoch": 3.506142506142506,
      "grad_norm": 0.32540553671935935,
      "learning_rate": 6.023459200347964e-05,
      "loss": 0.0738,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03626057878136635,
      "step": 715,
      "valid_targets_mean": 1160.9,
      "valid_targets_min": 682
    },
    {
      "epoch": 3.5307125307125307,
      "grad_norm": 0.3555455251156863,
      "learning_rate": 5.976044876937997e-05,
      "loss": 0.0723,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.038836769759655,
      "step": 720,
      "valid_targets_mean": 1545.9,
      "valid_targets_min": 865
    },
    {
      "epoch": 3.555282555282555,
      "grad_norm": 0.32715758611553336,
      "learning_rate": 5.9285391276561565e-05,
      "loss": 0.0728,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.036427922546863556,
      "step": 725,
      "valid_targets_mean": 1344.4,
      "valid_targets_min": 838
    },
    {
      "epoch": 3.57985257985258,
      "grad_norm": 0.33795073934789194,
      "learning_rate": 5.8809464023537265e-05,
      "loss": 0.0742,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.038653090596199036,
      "step": 730,
      "valid_targets_mean": 1377.6,
      "valid_targets_min": 732
    },
    {
      "epoch": 3.6044226044226044,
      "grad_norm": 0.3213417450855897,
      "learning_rate": 5.83327115902901e-05,
      "loss": 0.0742,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03317331522703171,
      "step": 735,
      "valid_targets_mean": 1256.9,
      "valid_targets_min": 825
    },
    {
      "epoch": 3.628992628992629,
      "grad_norm": 0.35052676334106075,
      "learning_rate": 5.785517863409752e-05,
      "loss": 0.0734,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.035290129482746124,
      "step": 740,
      "valid_targets_mean": 1192.2,
      "valid_targets_min": 457
    },
    {
      "epoch": 3.6535626535626538,
      "grad_norm": 0.3388956895842398,
      "learning_rate": 5.737690988534836e-05,
      "loss": 0.0729,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03399132192134857,
      "step": 745,
      "valid_targets_mean": 1213.5,
      "valid_targets_min": 527
    },
    {
      "epoch": 3.678132678132678,
      "grad_norm": 0.3669302957387778,
      "learning_rate": 5.689795014335296e-05,
      "loss": 0.0748,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.038341738283634186,
      "step": 750,
      "valid_targets_mean": 1240.8,
      "valid_targets_min": 624
    },
    {
      "epoch": 3.7027027027027026,
      "grad_norm": 0.3459031323835537,
      "learning_rate": 5.6418344272146816e-05,
      "loss": 0.0746,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03371729701757431,
      "step": 755,
      "valid_targets_mean": 1229.1,
      "valid_targets_min": 773
    },
    {
      "epoch": 3.7272727272727275,
      "grad_norm": 0.3246504376784696,
      "learning_rate": 5.593813719628819e-05,
      "loss": 0.0743,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03680756688117981,
      "step": 760,
      "valid_targets_mean": 1345.2,
      "valid_targets_min": 631
    },
    {
      "epoch": 3.751842751842752,
      "grad_norm": 0.31584812697899933,
      "learning_rate": 5.545737389664999e-05,
      "loss": 0.0745,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03372339904308319,
      "step": 765,
      "valid_targets_mean": 1635.1,
      "valid_targets_min": 1048
    },
    {
      "epoch": 3.7764127764127764,
      "grad_norm": 0.3254038680949618,
      "learning_rate": 5.4976099406206516e-05,
      "loss": 0.0714,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03635542094707489,
      "step": 770,
      "valid_targets_mean": 1342.9,
      "valid_targets_min": 734
    },
    {
      "epoch": 3.800982800982801,
      "grad_norm": 0.3462002699846917,
      "learning_rate": 5.449435880581513e-05,
      "loss": 0.0741,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03568622097373009,
      "step": 775,
      "valid_targets_mean": 1382.1,
      "valid_targets_min": 619
    },
    {
      "epoch": 3.8255528255528253,
      "grad_norm": 0.3267035574353153,
      "learning_rate": 5.401219721999364e-05,
      "loss": 0.0716,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03537577763199806,
      "step": 780,
      "valid_targets_mean": 1539.4,
      "valid_targets_min": 614
    },
    {
      "epoch": 3.85012285012285,
      "grad_norm": 0.38327734503562755,
      "learning_rate": 5.352965981269342e-05,
      "loss": 0.0726,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03703831881284714,
      "step": 785,
      "valid_targets_mean": 1188.2,
      "valid_targets_min": 600
    },
    {
      "epoch": 3.8746928746928746,
      "grad_norm": 0.334275811536087,
      "learning_rate": 5.304679178306894e-05,
      "loss": 0.0721,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.035076290369033813,
      "step": 790,
      "valid_targets_mean": 1480.4,
      "valid_targets_min": 475
    },
    {
      "epoch": 3.899262899262899,
      "grad_norm": 0.33191141500394544,
      "learning_rate": 5.2563638361244004e-05,
      "loss": 0.072,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03381187468767166,
      "step": 795,
      "valid_targets_mean": 1222.4,
      "valid_targets_min": 636
    },
    {
      "epoch": 3.923832923832924,
      "grad_norm": 0.3401577978749103,
      "learning_rate": 5.2080244804075e-05,
      "loss": 0.0716,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03545871376991272,
      "step": 800,
      "valid_targets_mean": 1154.8,
      "valid_targets_min": 704
    },
    {
      "epoch": 3.9484029484029484,
      "grad_norm": 0.32533812897018854,
      "learning_rate": 5.1596656390911756e-05,
      "loss": 0.0727,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.036433812230825424,
      "step": 805,
      "valid_targets_mean": 1252.9,
      "valid_targets_min": 758
    },
    {
      "epoch": 3.972972972972973,
      "grad_norm": 0.3090743671273376,
      "learning_rate": 5.111291841935619e-05,
      "loss": 0.0729,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.03902567923069,
      "step": 810,
      "valid_targets_mean": 1423.0,
      "valid_targets_min": 833
    },
    {
      "epoch": 3.9975429975429977,
      "grad_norm": 0.3823054079712023,
      "learning_rate": 5.0629076201019364e-05,
      "loss": 0.0749,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.053378112614154816,
      "step": 815,
      "valid_targets_mean": 1398.7,
      "valid_targets_min": 680
    },
    {
      "epoch": 4.019656019656019,
      "grad_norm": 0.27993141223279444,
      "learning_rate": 5.014517505727702e-05,
      "loss": 0.0453,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.018219877034425735,
      "step": 820,
      "valid_targets_mean": 1362.6,
      "valid_targets_min": 796
    },
    {
      "epoch": 4.044226044226044,
      "grad_norm": 0.3741991961810599,
      "learning_rate": 4.966126031502452e-05,
      "loss": 0.0413,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.019840583205223083,
      "step": 825,
      "valid_targets_mean": 1268.1,
      "valid_targets_min": 538
    },
    {
      "epoch": 4.068796068796069,
      "grad_norm": 0.30892475697952576,
      "learning_rate": 4.917737730243093e-05,
      "loss": 0.0384,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.02161870151758194,
      "step": 830,
      "valid_targets_mean": 1377.0,
      "valid_targets_min": 820
    },
    {
      "epoch": 4.093366093366093,
      "grad_norm": 0.3224276737619888,
      "learning_rate": 4.869357134469325e-05,
      "loss": 0.039,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.021871669217944145,
      "step": 835,
      "valid_targets_mean": 1343.4,
      "valid_targets_min": 711
    },
    {
      "epoch": 4.117936117936118,
      "grad_norm": 0.31835589619396676,
      "learning_rate": 4.820988775979074e-05,
      "loss": 0.0415,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.020772669464349747,
      "step": 840,
      "valid_targets_mean": 1411.6,
      "valid_targets_min": 697
    },
    {
      "epoch": 4.142506142506143,
      "grad_norm": 0.3185476210032366,
      "learning_rate": 4.772637185424005e-05,
      "loss": 0.0401,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.017333239316940308,
      "step": 845,
      "valid_targets_mean": 1268.7,
      "valid_targets_min": 848
    },
    {
      "epoch": 4.167076167076167,
      "grad_norm": 0.3080951677636193,
      "learning_rate": 4.724306891885134e-05,
      "loss": 0.0402,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.017767902463674545,
      "step": 850,
      "valid_targets_mean": 1320.8,
      "valid_targets_min": 628
    },
    {
      "epoch": 4.191646191646192,
      "grad_norm": 0.3045852265574028,
      "learning_rate": 4.6760024224485915e-05,
      "loss": 0.0412,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.020958252251148224,
      "step": 855,
      "valid_targets_mean": 1377.7,
      "valid_targets_min": 874
    },
    {
      "epoch": 4.216216216216216,
      "grad_norm": 0.33276273204262496,
      "learning_rate": 4.627728301781569e-05,
      "loss": 0.04,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.022474341094493866,
      "step": 860,
      "valid_targets_mean": 1423.2,
      "valid_targets_min": 840
    },
    {
      "epoch": 4.240786240786241,
      "grad_norm": 0.3173316619927852,
      "learning_rate": 4.5794890517084995e-05,
      "loss": 0.0404,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.021615831181406975,
      "step": 865,
      "valid_targets_mean": 1378.6,
      "valid_targets_min": 586
    },
    {
      "epoch": 4.2653562653562656,
      "grad_norm": 0.3367095693491688,
      "learning_rate": 4.531289190787493e-05,
      "loss": 0.0399,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.022174552083015442,
      "step": 870,
      "valid_targets_mean": 1354.9,
      "valid_targets_min": 613
    },
    {
      "epoch": 4.2899262899262895,
      "grad_norm": 0.31236014495985,
      "learning_rate": 4.483133233887093e-05,
      "loss": 0.0416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.02302752062678337,
      "step": 875,
      "valid_targets_mean": 1393.2,
      "valid_targets_min": 680
    },
    {
      "epoch": 4.314496314496314,
      "grad_norm": 0.3316531504924653,
      "learning_rate": 4.4350256917633585e-05,
      "loss": 0.0405,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.021897803992033005,
      "step": 880,
      "valid_targets_mean": 1357.8,
      "valid_targets_min": 709
    },
    {
      "epoch": 4.339066339066339,
      "grad_norm": 0.31737769102944213,
      "learning_rate": 4.386971070637354e-05,
      "loss": 0.0419,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.02140195667743683,
      "step": 885,
      "valid_targets_mean": 1482.6,
      "valid_targets_min": 957
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.3156768078020588,
      "learning_rate": 4.338973871773045e-05,
      "loss": 0.0404,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.019348224624991417,
      "step": 890,
      "valid_targets_mean": 1290.9,
      "valid_targets_min": 598
    },
    {
      "epoch": 4.388206388206388,
      "grad_norm": 0.32838758696752823,
      "learning_rate": 4.291038591055668e-05,
      "loss": 0.0391,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.018940959125757217,
      "step": 895,
      "valid_targets_mean": 1317.6,
      "valid_targets_min": 652
    },
    {
      "epoch": 4.412776412776413,
      "grad_norm": 0.3483390290744793,
      "learning_rate": 4.243169718570606e-05,
      "loss": 0.0419,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.022482123225927353,
      "step": 900,
      "valid_targets_mean": 1327.1,
      "valid_targets_min": 650
    },
    {
      "epoch": 4.437346437346437,
      "grad_norm": 0.31185423077558533,
      "learning_rate": 4.195371738182796e-05,
      "loss": 0.0388,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.01987990364432335,
      "step": 905,
      "valid_targets_mean": 1322.4,
      "valid_targets_min": 681
    },
    {
      "epoch": 4.461916461916462,
      "grad_norm": 0.3520431767140775,
      "learning_rate": 4.147649127116735e-05,
      "loss": 0.0416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.021981684491038322,
      "step": 910,
      "valid_targets_mean": 1278.7,
      "valid_targets_min": 651
    },
    {
      "epoch": 4.486486486486487,
      "grad_norm": 0.3139898395853098,
      "learning_rate": 4.1000063555370894e-05,
      "loss": 0.0416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.020710209384560585,
      "step": 915,
      "valid_targets_mean": 1371.6,
      "valid_targets_min": 501
    },
    {
      "epoch": 4.511056511056511,
      "grad_norm": 0.34028718296899796,
      "learning_rate": 4.052447886129986e-05,
      "loss": 0.0413,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.022445928305387497,
      "step": 920,
      "valid_targets_mean": 1329.0,
      "valid_targets_min": 686
    },
    {
      "epoch": 4.535626535626536,
      "grad_norm": 0.34533064643329925,
      "learning_rate": 4.004978173684988e-05,
      "loss": 0.0421,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.018675517290830612,
      "step": 925,
      "valid_targets_mean": 1227.1,
      "valid_targets_min": 703
    },
    {
      "epoch": 4.560196560196561,
      "grad_norm": 0.3234418433633803,
      "learning_rate": 3.957601664677816e-05,
      "loss": 0.0409,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.018068864941596985,
      "step": 930,
      "valid_targets_mean": 1164.3,
      "valid_targets_min": 671
    },
    {
      "epoch": 4.584766584766585,
      "grad_norm": 0.32817885450102535,
      "learning_rate": 3.910322796853848e-05,
      "loss": 0.0413,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.023279931396245956,
      "step": 935,
      "valid_targets_mean": 1372.6,
      "valid_targets_min": 770
    },
    {
      "epoch": 4.6093366093366095,
      "grad_norm": 0.31227498772612844,
      "learning_rate": 3.86314599881244e-05,
      "loss": 0.041,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.021624624729156494,
      "step": 940,
      "valid_targets_mean": 1521.3,
      "valid_targets_min": 555
    },
    {
      "epoch": 4.6339066339066335,
      "grad_norm": 0.32927359413012675,
      "learning_rate": 3.816075689592095e-05,
      "loss": 0.0409,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.02119259163737297,
      "step": 945,
      "valid_targets_mean": 1311.4,
      "valid_targets_min": 670
    },
    {
      "epoch": 4.658476658476658,
      "grad_norm": 0.3264412276653931,
      "learning_rate": 3.7691162782565383e-05,
      "loss": 0.0378,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.019723929464817047,
      "step": 950,
      "valid_targets_mean": 1315.0,
      "valid_targets_min": 576
    },
    {
      "epoch": 4.683046683046683,
      "grad_norm": 0.3109817640497461,
      "learning_rate": 3.7222721634817146e-05,
      "loss": 0.0405,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.019970854744315147,
      "step": 955,
      "valid_targets_mean": 1293.0,
      "valid_targets_min": 734
    },
    {
      "epoch": 4.707616707616707,
      "grad_norm": 0.30543197648210796,
      "learning_rate": 3.675547733143776e-05,
      "loss": 0.0389,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.017612136900424957,
      "step": 960,
      "valid_targets_mean": 1288.9,
      "valid_targets_min": 762
    },
    {
      "epoch": 4.732186732186732,
      "grad_norm": 0.3211505739622534,
      "learning_rate": 3.628947363908058e-05,
      "loss": 0.0399,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.019393999129533768,
      "step": 965,
      "valid_targets_mean": 1484.4,
      "valid_targets_min": 863
    },
    {
      "epoch": 4.756756756756757,
      "grad_norm": 0.33268734426995283,
      "learning_rate": 3.582475420819129e-05,
      "loss": 0.0395,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.017341013997793198,
      "step": 970,
      "valid_targets_mean": 1231.2,
      "valid_targets_min": 625
    },
    {
      "epoch": 4.781326781326781,
      "grad_norm": 0.320969799122163,
      "learning_rate": 3.53613625689191e-05,
      "loss": 0.0399,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.01869683712720871,
      "step": 975,
      "valid_targets_mean": 1218.4,
      "valid_targets_min": 624
    },
    {
      "epoch": 4.805896805896806,
      "grad_norm": 0.3114776209556068,
      "learning_rate": 3.489934212703936e-05,
      "loss": 0.0393,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.018647339195013046,
      "step": 980,
      "valid_targets_mean": 1183.8,
      "valid_targets_min": 559
    },
    {
      "epoch": 4.830466830466831,
      "grad_norm": 0.33522401899380994,
      "learning_rate": 3.4438736159887665e-05,
      "loss": 0.0381,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.019779419526457787,
      "step": 985,
      "valid_targets_mean": 1327.4,
      "valid_targets_min": 639
    },
    {
      "epoch": 4.855036855036855,
      "grad_norm": 0.3382381310169906,
      "learning_rate": 3.3979587812306196e-05,
      "loss": 0.0381,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.01958424225449562,
      "step": 990,
      "valid_targets_mean": 1366.7,
      "valid_targets_min": 822
    },
    {
      "epoch": 4.87960687960688,
      "grad_norm": 0.3315405268055381,
      "learning_rate": 3.352194009260221e-05,
      "loss": 0.0401,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.018945585936307907,
      "step": 995,
      "valid_targets_mean": 1148.9,
      "valid_targets_min": 606
    },
    {
      "epoch": 4.9041769041769046,
      "grad_norm": 0.3115079992976071,
      "learning_rate": 3.306583586851956e-05,
      "loss": 0.0376,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.021324384957551956,
      "step": 1000,
      "valid_targets_mean": 1306.4,
      "valid_targets_min": 571
    },
    {
      "epoch": 4.9287469287469285,
      "grad_norm": 0.3078542084967762,
      "learning_rate": 3.26113178632233e-05,
      "loss": 0.0416,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.02121821790933609,
      "step": 1005,
      "valid_targets_mean": 1202.2,
      "valid_targets_min": 576
    },
    {
      "epoch": 4.953316953316953,
      "grad_norm": 0.3017240699598429,
      "learning_rate": 3.215842865129773e-05,
      "loss": 0.0385,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.01800459809601307,
      "step": 1010,
      "valid_targets_mean": 1340.1,
      "valid_targets_min": 587
    },
    {
      "epoch": 4.977886977886978,
      "grad_norm": 0.3247961752433559,
      "learning_rate": 3.1707210654758556e-05,
      "loss": 0.038,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.01770874857902527,
      "step": 1015,
      "valid_targets_mean": 1250.6,
      "valid_targets_min": 632
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.36428797184528683,
      "learning_rate": 3.125770613907909e-05,
      "loss": 0.0364,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.027929894626140594,
      "step": 1020,
      "valid_targets_mean": 1413.3,
      "valid_targets_min": 603
    },
    {
      "epoch": 5.024570024570025,
      "grad_norm": 0.26133962940041444,
      "learning_rate": 3.08099572092314e-05,
      "loss": 0.0204,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010166799649596214,
      "step": 1025,
      "valid_targets_mean": 1211.6,
      "valid_targets_min": 668
    },
    {
      "epoch": 5.049140049140049,
      "grad_norm": 0.35013474614494267,
      "learning_rate": 3.0364005805742246e-05,
      "loss": 0.0206,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010145038366317749,
      "step": 1030,
      "valid_targets_mean": 1168.4,
      "valid_targets_min": 666
    },
    {
      "epoch": 5.073710073710074,
      "grad_norm": 0.282893581071777,
      "learning_rate": 2.9919893700764566e-05,
      "loss": 0.0202,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.01128455065190792,
      "step": 1035,
      "valid_targets_mean": 1338.1,
      "valid_targets_min": 741
    },
    {
      "epoch": 5.098280098280099,
      "grad_norm": 0.28533102889928075,
      "learning_rate": 2.9477662494164703e-05,
      "loss": 0.02,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010152880102396011,
      "step": 1040,
      "valid_targets_mean": 1345.1,
      "valid_targets_min": 564
    },
    {
      "epoch": 5.122850122850123,
      "grad_norm": 0.314478747850382,
      "learning_rate": 2.9037353609625695e-05,
      "loss": 0.021,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010847135446965694,
      "step": 1045,
      "valid_targets_mean": 1294.2,
      "valid_targets_min": 680
    },
    {
      "epoch": 5.1474201474201475,
      "grad_norm": 0.26706110402851246,
      "learning_rate": 2.8599008290767204e-05,
      "loss": 0.0193,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.009428557008504868,
      "step": 1050,
      "valid_targets_mean": 1452.9,
      "valid_targets_min": 725
    },
    {
      "epoch": 5.171990171990172,
      "grad_norm": 0.29886945041340757,
      "learning_rate": 2.8162667597282176e-05,
      "loss": 0.0204,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.012537685222923756,
      "step": 1055,
      "valid_targets_mean": 1264.2,
      "valid_targets_min": 669
    },
    {
      "epoch": 5.196560196560196,
      "grad_norm": 0.3020986480627111,
      "learning_rate": 2.7728372401090806e-05,
      "loss": 0.0197,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010673290118575096,
      "step": 1060,
      "valid_targets_mean": 1316.6,
      "valid_targets_min": 609
    },
    {
      "epoch": 5.221130221130221,
      "grad_norm": 0.2900262078972821,
      "learning_rate": 2.729616338251215e-05,
      "loss": 0.0198,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.009456290863454342,
      "step": 1065,
      "valid_targets_mean": 1233.6,
      "valid_targets_min": 736
    },
    {
      "epoch": 5.245700245700245,
      "grad_norm": 0.3084078147531761,
      "learning_rate": 2.686608102645347e-05,
      "loss": 0.0196,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.009841931983828545,
      "step": 1070,
      "valid_targets_mean": 1166.8,
      "valid_targets_min": 849
    },
    {
      "epoch": 5.27027027027027,
      "grad_norm": 0.299437980932929,
      "learning_rate": 2.6438165618618127e-05,
      "loss": 0.0203,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.011637433432042599,
      "step": 1075,
      "valid_targets_mean": 1208.6,
      "valid_targets_min": 700
    },
    {
      "epoch": 5.294840294840295,
      "grad_norm": 0.3077946376633561,
      "learning_rate": 2.6012457241731986e-05,
      "loss": 0.0194,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.01187128946185112,
      "step": 1080,
      "valid_targets_mean": 1109.1,
      "valid_targets_min": 577
    },
    {
      "epoch": 5.319410319410319,
      "grad_norm": 0.30109678654303274,
      "learning_rate": 2.5588995771788942e-05,
      "loss": 0.0213,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.01099569071084261,
      "step": 1085,
      "valid_targets_mean": 1176.6,
      "valid_targets_min": 697
    },
    {
      "epoch": 5.343980343980344,
      "grad_norm": 0.2895876653364538,
      "learning_rate": 2.516782087431565e-05,
      "loss": 0.0198,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.00905087310820818,
      "step": 1090,
      "valid_targets_mean": 1213.7,
      "valid_targets_min": 680
    },
    {
      "epoch": 5.368550368550369,
      "grad_norm": 0.3103618365984646,
      "learning_rate": 2.474897200065611e-05,
      "loss": 0.0198,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.012017014436423779,
      "step": 1095,
      "valid_targets_mean": 1355.6,
      "valid_targets_min": 841
    },
    {
      "epoch": 5.393120393120393,
      "grad_norm": 0.281975216305729,
      "learning_rate": 2.433248838427628e-05,
      "loss": 0.0206,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.011380782350897789,
      "step": 1100,
      "valid_targets_mean": 1249.9,
      "valid_targets_min": 690
    },
    {
      "epoch": 5.417690417690418,
      "grad_norm": 0.25746102969425927,
      "learning_rate": 2.3918409037089112e-05,
      "loss": 0.0187,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.008737726137042046,
      "step": 1105,
      "valid_targets_mean": 1287.6,
      "valid_targets_min": 650
    },
    {
      "epoch": 5.442260442260443,
      "grad_norm": 0.3169046454091352,
      "learning_rate": 2.3506772745800238e-05,
      "loss": 0.0192,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010114330798387527,
      "step": 1110,
      "valid_targets_mean": 1159.0,
      "valid_targets_min": 619
    },
    {
      "epoch": 5.466830466830467,
      "grad_norm": 0.2540569088631003,
      "learning_rate": 2.309761806827489e-05,
      "loss": 0.0183,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.008340870961546898,
      "step": 1115,
      "valid_targets_mean": 1512.6,
      "valid_targets_min": 864
    },
    {
      "epoch": 5.4914004914004915,
      "grad_norm": 0.26922436587552395,
      "learning_rate": 2.2690983329926157e-05,
      "loss": 0.019,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.008892050012946129,
      "step": 1120,
      "valid_targets_mean": 1536.9,
      "valid_targets_min": 789
    },
    {
      "epoch": 5.515970515970516,
      "grad_norm": 0.3138051935360876,
      "learning_rate": 2.228690662012514e-05,
      "loss": 0.0198,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.011595183052122593,
      "step": 1125,
      "valid_targets_mean": 1244.7,
      "valid_targets_min": 651
    },
    {
      "epoch": 5.54054054054054,
      "grad_norm": 0.3117819215203409,
      "learning_rate": 2.1885425788633e-05,
      "loss": 0.0201,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010435843840241432,
      "step": 1130,
      "valid_targets_mean": 1302.9,
      "valid_targets_min": 711
    },
    {
      "epoch": 5.565110565110565,
      "grad_norm": 0.3069592336125717,
      "learning_rate": 2.1486578442055672e-05,
      "loss": 0.02,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010668829083442688,
      "step": 1135,
      "valid_targets_mean": 1159.2,
      "valid_targets_min": 648
    },
    {
      "epoch": 5.58968058968059,
      "grad_norm": 0.31656524373785344,
      "learning_rate": 2.1090401940321212e-05,
      "loss": 0.0181,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.011099237948656082,
      "step": 1140,
      "valid_targets_mean": 1232.4,
      "valid_targets_min": 682
    },
    {
      "epoch": 5.614250614250614,
      "grad_norm": 0.2670591117298401,
      "learning_rate": 2.0696933393180397e-05,
      "loss": 0.0176,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010421942919492722,
      "step": 1145,
      "valid_targets_mean": 1465.6,
      "valid_targets_min": 693
    },
    {
      "epoch": 5.638820638820639,
      "grad_norm": 0.3182550958234908,
      "learning_rate": 2.0306209656730523e-05,
      "loss": 0.0187,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.010414216667413712,
      "step": 1150,
      "valid_targets_mean": 1375.6,
      "valid_targets_min": 663
    },
    {
      "epoch": 5.663390663390663,
      "grad_norm": 0.2662439452158416,
      "learning_rate": 1.991826732996319e-05,
      "loss": 0.0184,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.009408965706825256,
      "step": 1155,
      "valid_targets_mean": 1455.5,
      "valid_targets_min": 857
    },
    {
      "epoch": 5.687960687960688,
      "grad_norm": 0.25849619635330695,
      "learning_rate": 1.9533142751336126e-05,
      "loss": 0.0182,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.008321743458509445,
      "step": 1160,
      "valid_targets_mean": 1318.8,
      "valid_targets_min": 648
    },
    {
      "epoch": 5.712530712530713,
      "grad_norm": 0.2421276566574057,
      "learning_rate": 1.915087199536925e-05,
      "loss": 0.0185,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.007764388341456652,
      "step": 1165,
      "valid_targets_mean": 1301.2,
      "valid_targets_min": 673
    },
    {
      "epoch": 5.737100737100737,
      "grad_norm": 0.2909298663388158,
      "learning_rate": 1.8771490869265686e-05,
      "loss": 0.0181,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.009419236332178116,
      "step": 1170,
      "valid_targets_mean": 1192.9,
      "valid_targets_min": 690
    },
    {
      "epoch": 5.761670761670762,
      "grad_norm": 0.26713609394086596,
      "learning_rate": 1.839503490955763e-05,
      "loss": 0.0176,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.008326014503836632,
      "step": 1175,
      "valid_targets_mean": 1488.6,
      "valid_targets_min": 647
    },
    {
      "epoch": 5.7862407862407865,
      "grad_norm": 0.26467576781241553,
      "learning_rate": 1.802153937877777e-05,
      "loss": 0.0183,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.008647764101624489,
      "step": 1180,
      "valid_targets_mean": 1414.1,
      "valid_targets_min": 559
    },
    {
      "epoch": 5.8108108108108105,
      "grad_norm": 0.26641667371207894,
      "learning_rate": 1.7651039262156126e-05,
      "loss": 0.0173,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.008846086449921131,
      "step": 1185,
      "valid_targets_mean": 1301.8,
      "valid_targets_min": 709
    },
    {
      "epoch": 5.835380835380835,
      "grad_norm": 0.31650104270109913,
      "learning_rate": 1.728356926434306e-05,
      "loss": 0.0178,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.009932863526046276,
      "step": 1190,
      "valid_targets_mean": 1253.7,
      "valid_targets_min": 728
    },
    {
      "epoch": 5.85995085995086,
      "grad_norm": 0.2836541956422589,
      "learning_rate": 1.6919163806158455e-05,
      "loss": 0.0184,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.008826677687466145,
      "step": 1195,
      "valid_targets_mean": 1219.0,
      "valid_targets_min": 563
    },
    {
      "epoch": 5.884520884520884,
      "grad_norm": 0.2943453272730564,
      "learning_rate": 1.655785702136764e-05,
      "loss": 0.0177,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.009762442670762539,
      "step": 1200,
      "valid_targets_mean": 1260.2,
      "valid_targets_min": 783
    },
    {
      "epoch": 5.909090909090909,
      "grad_norm": 0.27725462162946896,
      "learning_rate": 1.6199682753483926e-05,
      "loss": 0.0181,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.007736817002296448,
      "step": 1205,
      "valid_targets_mean": 1286.1,
      "valid_targets_min": 679
    },
    {
      "epoch": 5.933660933660933,
      "grad_norm": 0.2823492995999207,
      "learning_rate": 1.584467455259861e-05,
      "loss": 0.0213,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.00895984098315239,
      "step": 1210,
      "valid_targets_mean": 1235.6,
      "valid_targets_min": 770
    },
    {
      "epoch": 5.958230958230958,
      "grad_norm": 0.274768095515548,
      "learning_rate": 1.5492865672238276e-05,
      "loss": 0.0178,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.009178702719509602,
      "step": 1215,
      "valid_targets_mean": 1213.4,
      "valid_targets_min": 575
    },
    {
      "epoch": 5.982800982800983,
      "grad_norm": 0.2571522327732023,
      "learning_rate": 1.5144289066250045e-05,
      "loss": 0.0178,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.008877310901880264,
      "step": 1220,
      "valid_targets_mean": 1334.7,
      "valid_targets_min": 751
    },
    {
      "epoch": 6.004914004914005,
      "grad_norm": 0.16174171115393154,
      "learning_rate": 1.479897738571468e-05,
      "loss": 0.0152,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004827571101486683,
      "step": 1225,
      "valid_targets_mean": 1169.8,
      "valid_targets_min": 582
    },
    {
      "epoch": 6.0294840294840295,
      "grad_norm": 0.18569722400832173,
      "learning_rate": 1.4456962975888216e-05,
      "loss": 0.0078,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035782852210104465,
      "step": 1230,
      "valid_targets_mean": 1211.8,
      "valid_targets_min": 607
    },
    {
      "epoch": 6.054054054054054,
      "grad_norm": 0.21491424856633037,
      "learning_rate": 1.4118277873172208e-05,
      "loss": 0.0081,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004247656557708979,
      "step": 1235,
      "valid_targets_mean": 1248.5,
      "valid_targets_min": 639
    },
    {
      "epoch": 6.078624078624078,
      "grad_norm": 0.25016395772751115,
      "learning_rate": 1.378295380211289e-05,
      "loss": 0.0082,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0046090735122561455,
      "step": 1240,
      "valid_targets_mean": 1469.5,
      "valid_targets_min": 899
    },
    {
      "epoch": 6.103194103194103,
      "grad_norm": 0.22993854572734307,
      "learning_rate": 1.3451022172429495e-05,
      "loss": 0.0086,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004097792319953442,
      "step": 1245,
      "valid_targets_mean": 1390.2,
      "valid_targets_min": 845
    },
    {
      "epoch": 6.127764127764128,
      "grad_norm": 0.20768384003216256,
      "learning_rate": 1.3122514076072163e-05,
      "loss": 0.0077,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0044163367711007595,
      "step": 1250,
      "valid_targets_mean": 1341.2,
      "valid_targets_min": 897
    },
    {
      "epoch": 6.152334152334152,
      "grad_norm": 0.22806257458798243,
      "learning_rate": 1.2797460284309532e-05,
      "loss": 0.008,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003973452374339104,
      "step": 1255,
      "valid_targets_mean": 1167.9,
      "valid_targets_min": 715
    },
    {
      "epoch": 6.176904176904177,
      "grad_norm": 0.18667243280171153,
      "learning_rate": 1.247589124484646e-05,
      "loss": 0.0076,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003657914698123932,
      "step": 1260,
      "valid_targets_mean": 1442.6,
      "valid_targets_min": 636
    },
    {
      "epoch": 6.201474201474202,
      "grad_norm": 0.23820555371148794,
      "learning_rate": 1.2157837078971928e-05,
      "loss": 0.008,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003562439698725939,
      "step": 1265,
      "valid_targets_mean": 1425.5,
      "valid_targets_min": 755
    },
    {
      "epoch": 6.226044226044226,
      "grad_norm": 0.26784606697716024,
      "learning_rate": 1.1843327578737612e-05,
      "loss": 0.0084,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.005037506110966206,
      "step": 1270,
      "valid_targets_mean": 1198.7,
      "valid_targets_min": 726
    },
    {
      "epoch": 6.250614250614251,
      "grad_norm": 0.21603907352417645,
      "learning_rate": 1.1532392204167275e-05,
      "loss": 0.0075,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0038490337319672108,
      "step": 1275,
      "valid_targets_mean": 1141.4,
      "valid_targets_min": 698
    },
    {
      "epoch": 6.275184275184275,
      "grad_norm": 0.2346178380054373,
      "learning_rate": 1.1225060080497257e-05,
      "loss": 0.0081,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004704365506768227,
      "step": 1280,
      "valid_targets_mean": 1300.5,
      "valid_targets_min": 577
    },
    {
      "epoch": 6.2997542997543,
      "grad_norm": 0.18637031994574188,
      "learning_rate": 1.092135999544831e-05,
      "loss": 0.0075,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0033987853676080704,
      "step": 1285,
      "valid_targets_mean": 1284.4,
      "valid_targets_min": 760
    },
    {
      "epoch": 6.324324324324325,
      "grad_norm": 0.24995090437047596,
      "learning_rate": 1.0621320396529056e-05,
      "loss": 0.0084,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004618373699486256,
      "step": 1290,
      "valid_targets_mean": 1206.5,
      "valid_targets_min": 633
    },
    {
      "epoch": 6.348894348894349,
      "grad_norm": 0.218872542773124,
      "learning_rate": 1.0324969388371364e-05,
      "loss": 0.0077,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004193416330963373,
      "step": 1295,
      "valid_targets_mean": 1309.2,
      "valid_targets_min": 728
    },
    {
      "epoch": 6.3734643734643734,
      "grad_norm": 0.19603164940260448,
      "learning_rate": 1.0032334730097715e-05,
      "loss": 0.0078,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003843929385766387,
      "step": 1300,
      "valid_targets_mean": 1404.2,
      "valid_targets_min": 675
    },
    {
      "epoch": 6.398034398034398,
      "grad_norm": 0.20046639841749483,
      "learning_rate": 9.743443832721055e-06,
      "loss": 0.0079,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004076910670846701,
      "step": 1305,
      "valid_targets_mean": 1295.1,
      "valid_targets_min": 601
    },
    {
      "epoch": 6.422604422604422,
      "grad_norm": 0.21634516207911922,
      "learning_rate": 9.458323756577264e-06,
      "loss": 0.0076,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035616583190858364,
      "step": 1310,
      "valid_targets_mean": 1419.7,
      "valid_targets_min": 820
    },
    {
      "epoch": 6.447174447174447,
      "grad_norm": 0.1920680703820834,
      "learning_rate": 9.17700120879031e-06,
      "loss": 0.007,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003503095591440797,
      "step": 1315,
      "valid_targets_mean": 1528.4,
      "valid_targets_min": 964
    },
    {
      "epoch": 6.471744471744472,
      "grad_norm": 0.22880019301249582,
      "learning_rate": 8.899502540770688e-06,
      "loss": 0.008,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0037546579260379076,
      "step": 1320,
      "valid_targets_mean": 1280.2,
      "valid_targets_min": 633
    },
    {
      "epoch": 6.496314496314496,
      "grad_norm": 0.2514546590455702,
      "learning_rate": 8.625853745747048e-06,
      "loss": 0.0074,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004576146602630615,
      "step": 1325,
      "valid_targets_mean": 1257.1,
      "valid_targets_min": 575
    },
    {
      "epoch": 6.520884520884521,
      "grad_norm": 0.2554072646178069,
      "learning_rate": 8.35608045633145e-06,
      "loss": 0.0073,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0029603377915918827,
      "step": 1330,
      "valid_targets_mean": 1260.9,
      "valid_targets_min": 694
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.215953034629028,
      "learning_rate": 8.090207942118333e-06,
      "loss": 0.0076,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.00399182690307498,
      "step": 1335,
      "valid_targets_mean": 1512.6,
      "valid_targets_min": 622
    },
    {
      "epoch": 6.57002457002457,
      "grad_norm": 0.2388743990811728,
      "learning_rate": 7.82826110731752e-06,
      "loss": 0.0074,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004739910364151001,
      "step": 1340,
      "valid_targets_mean": 1261.9,
      "valid_targets_min": 550
    },
    {
      "epoch": 6.594594594594595,
      "grad_norm": 0.19693613080448677,
      "learning_rate": 7.570264488421447e-06,
      "loss": 0.0075,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0029330942779779434,
      "step": 1345,
      "valid_targets_mean": 1331.1,
      "valid_targets_min": 786
    },
    {
      "epoch": 6.61916461916462,
      "grad_norm": 0.21706119299314372,
      "learning_rate": 7.3162422519068966e-06,
      "loss": 0.0076,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0039455159567296505,
      "step": 1350,
      "valid_targets_mean": 1311.2,
      "valid_targets_min": 614
    },
    {
      "epoch": 6.643734643734644,
      "grad_norm": 0.169159406905177,
      "learning_rate": 7.066218191971219e-06,
      "loss": 0.0073,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0025496738962829113,
      "step": 1355,
      "valid_targets_mean": 1424.5,
      "valid_targets_min": 732
    },
    {
      "epoch": 6.6683046683046685,
      "grad_norm": 0.2122964661186571,
      "learning_rate": 6.820215728303625e-06,
      "loss": 0.0071,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0036278744228184223,
      "step": 1360,
      "valid_targets_mean": 1245.1,
      "valid_targets_min": 629
    },
    {
      "epoch": 6.6928746928746925,
      "grad_norm": 0.20634084259782276,
      "learning_rate": 6.578257903891427e-06,
      "loss": 0.0066,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003295808332040906,
      "step": 1365,
      "valid_targets_mean": 1429.2,
      "valid_targets_min": 730
    },
    {
      "epoch": 6.717444717444717,
      "grad_norm": 0.21991786675018984,
      "learning_rate": 6.34036738286165e-06,
      "loss": 0.0071,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0043110596016049385,
      "step": 1370,
      "valid_targets_mean": 1328.5,
      "valid_targets_min": 693
    },
    {
      "epoch": 6.742014742014742,
      "grad_norm": 0.24788989146170717,
      "learning_rate": 6.106566448358025e-06,
      "loss": 0.0071,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004148700274527073,
      "step": 1375,
      "valid_targets_mean": 1244.7,
      "valid_targets_min": 796
    },
    {
      "epoch": 6.766584766584766,
      "grad_norm": 0.20977102960147215,
      "learning_rate": 5.8768770004537894e-06,
      "loss": 0.007,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0038808349054306746,
      "step": 1380,
      "valid_targets_mean": 1229.1,
      "valid_targets_min": 679
    },
    {
      "epoch": 6.791154791154791,
      "grad_norm": 0.18128751410719413,
      "learning_rate": 5.65132055410027e-06,
      "loss": 0.0065,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003224300919100642,
      "step": 1385,
      "valid_targets_mean": 1343.8,
      "valid_targets_min": 740
    },
    {
      "epoch": 6.815724815724816,
      "grad_norm": 0.2243474366269306,
      "learning_rate": 5.429918237111642e-06,
      "loss": 0.0059,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0031104448717087507,
      "step": 1390,
      "valid_targets_mean": 1281.1,
      "valid_targets_min": 508
    },
    {
      "epoch": 6.84029484029484,
      "grad_norm": 0.22158595978990048,
      "learning_rate": 5.21269078818582e-06,
      "loss": 0.0067,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0033817810472100973,
      "step": 1395,
      "valid_targets_mean": 1257.1,
      "valid_targets_min": 840
    },
    {
      "epoch": 6.864864864864865,
      "grad_norm": 0.23250133634103248,
      "learning_rate": 4.999658554961917e-06,
      "loss": 0.0094,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004702032543718815,
      "step": 1400,
      "valid_targets_mean": 1354.7,
      "valid_targets_min": 555
    },
    {
      "epoch": 6.88943488943489,
      "grad_norm": 0.22241489932962594,
      "learning_rate": 4.790841492114256e-06,
      "loss": 0.0073,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003928256221115589,
      "step": 1405,
      "valid_targets_mean": 1285.6,
      "valid_targets_min": 773
    },
    {
      "epoch": 6.914004914004914,
      "grad_norm": 0.22390574620767395,
      "learning_rate": 4.586259159483286e-06,
      "loss": 0.0063,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003446041140705347,
      "step": 1410,
      "valid_targets_mean": 1231.8,
      "valid_targets_min": 697
    },
    {
      "epoch": 6.938574938574939,
      "grad_norm": 0.23327794938321245,
      "learning_rate": 4.385930720243314e-06,
      "loss": 0.0067,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0034866356290876865,
      "step": 1415,
      "valid_targets_mean": 1175.0,
      "valid_targets_min": 801
    },
    {
      "epoch": 6.963144963144963,
      "grad_norm": 0.22248429253719407,
      "learning_rate": 4.189874939107574e-06,
      "loss": 0.007,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0036289298441261053,
      "step": 1420,
      "valid_targets_mean": 1280.1,
      "valid_targets_min": 671
    },
    {
      "epoch": 6.987714987714988,
      "grad_norm": 0.18240933384590993,
      "learning_rate": 3.998110180570525e-06,
      "loss": 0.0061,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0025913119316101074,
      "step": 1425,
      "valid_targets_mean": 1412.8,
      "valid_targets_min": 613
    },
    {
      "epoch": 7.00982800982801,
      "grad_norm": 0.11518082939815999,
      "learning_rate": 3.810654407187636e-06,
      "loss": 0.005,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001377458916977048,
      "step": 1430,
      "valid_targets_mean": 1409.6,
      "valid_targets_min": 800
    },
    {
      "epoch": 7.034398034398034,
      "grad_norm": 0.11366256392530506,
      "learning_rate": 3.6275251778928487e-06,
      "loss": 0.0029,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0015742455143481493,
      "step": 1435,
      "valid_targets_mean": 1455.5,
      "valid_targets_min": 999
    },
    {
      "epoch": 7.058968058968059,
      "grad_norm": 0.10653619054329262,
      "learning_rate": 3.4487396463538215e-06,
      "loss": 0.0027,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001220267964527011,
      "step": 1440,
      "valid_targets_mean": 1301.7,
      "valid_targets_min": 686
    },
    {
      "epoch": 7.083538083538084,
      "grad_norm": 0.11194152489721383,
      "learning_rate": 3.2743145593652047e-06,
      "loss": 0.003,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0016661153640598059,
      "step": 1445,
      "valid_targets_mean": 1376.6,
      "valid_targets_min": 725
    },
    {
      "epoch": 7.108108108108108,
      "grad_norm": 0.12343103375814024,
      "learning_rate": 3.1042662552798975e-06,
      "loss": 0.0028,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0012464872561395168,
      "step": 1450,
      "valid_targets_mean": 1389.7,
      "valid_targets_min": 626
    },
    {
      "epoch": 7.132678132678133,
      "grad_norm": 0.12753242419982108,
      "learning_rate": 2.9386106624786804e-06,
      "loss": 0.0027,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0013661307748407125,
      "step": 1455,
      "valid_targets_mean": 1296.8,
      "valid_targets_min": 716
    },
    {
      "epoch": 7.157248157248158,
      "grad_norm": 0.14659785771116587,
      "learning_rate": 2.7773632978781936e-06,
      "loss": 0.0026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0012009042548015714,
      "step": 1460,
      "valid_targets_mean": 1220.9,
      "valid_targets_min": 639
    },
    {
      "epoch": 7.181818181818182,
      "grad_norm": 0.11218152706881185,
      "learning_rate": 2.620539265477512e-06,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0011191873345524073,
      "step": 1465,
      "valid_targets_mean": 1306.1,
      "valid_targets_min": 873
    },
    {
      "epoch": 7.2063882063882065,
      "grad_norm": 0.12983333466333855,
      "learning_rate": 2.468153254943284e-06,
      "loss": 0.0026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0014806350227445364,
      "step": 1470,
      "valid_targets_mean": 1267.2,
      "valid_targets_min": 700
    },
    {
      "epoch": 7.2309582309582305,
      "grad_norm": 0.08990504914273968,
      "learning_rate": 2.32021954023382e-06,
      "loss": 0.0028,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001308899256400764,
      "step": 1475,
      "valid_targets_mean": 1533.7,
      "valid_targets_min": 779
    },
    {
      "epoch": 7.255528255528255,
      "grad_norm": 0.14187065725212764,
      "learning_rate": 2.1767519782620095e-06,
      "loss": 0.0028,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001910665538161993,
      "step": 1480,
      "valid_targets_mean": 1316.4,
      "valid_targets_min": 762
    },
    {
      "epoch": 7.28009828009828,
      "grad_norm": 0.13580002508297487,
      "learning_rate": 2.0377640075973926e-06,
      "loss": 0.0028,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0013501509092748165,
      "step": 1485,
      "valid_targets_mean": 1464.1,
      "valid_targets_min": 734
    },
    {
      "epoch": 7.304668304668304,
      "grad_norm": 0.11703590507552636,
      "learning_rate": 1.903268647207329e-06,
      "loss": 0.0024,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.000767023244407028,
      "step": 1490,
      "valid_targets_mean": 1434.1,
      "valid_targets_min": 738
    },
    {
      "epoch": 7.329238329238329,
      "grad_norm": 0.1259216278943979,
      "learning_rate": 1.7732784952375236e-06,
      "loss": 0.0027,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0014803120866417885,
      "step": 1495,
      "valid_targets_mean": 1378.4,
      "valid_targets_min": 797
    },
    {
      "epoch": 7.353808353808354,
      "grad_norm": 0.11355256723687553,
      "learning_rate": 1.6478057278319914e-06,
      "loss": 0.0026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0012061558663845062,
      "step": 1500,
      "valid_targets_mean": 1271.9,
      "valid_targets_min": 720
    },
    {
      "epoch": 7.378378378378378,
      "grad_norm": 0.11340763399720942,
      "learning_rate": 1.5268620979924986e-06,
      "loss": 0.0026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0010582938557490706,
      "step": 1505,
      "valid_targets_mean": 1201.9,
      "valid_targets_min": 801
    },
    {
      "epoch": 7.402948402948403,
      "grad_norm": 0.12412291642290835,
      "learning_rate": 1.4104589344776542e-06,
      "loss": 0.0024,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0015159003669396043,
      "step": 1510,
      "valid_targets_mean": 1675.2,
      "valid_targets_min": 937
    },
    {
      "epoch": 7.427518427518428,
      "grad_norm": 0.13669679018868425,
      "learning_rate": 1.2986071407417533e-06,
      "loss": 0.0024,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001106357667595148,
      "step": 1515,
      "valid_targets_mean": 1426.0,
      "valid_targets_min": 592
    },
    {
      "epoch": 7.452088452088452,
      "grad_norm": 0.12301018603265855,
      "learning_rate": 1.1913171939134659e-06,
      "loss": 0.0024,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001434761332347989,
      "step": 1520,
      "valid_targets_mean": 1361.7,
      "valid_targets_min": 659
    },
    {
      "epoch": 7.476658476658477,
      "grad_norm": 0.11562603789453654,
      "learning_rate": 1.0885991438144448e-06,
      "loss": 0.0028,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0013514563906937838,
      "step": 1525,
      "valid_targets_mean": 1447.6,
      "valid_targets_min": 959
    },
    {
      "epoch": 7.501228501228502,
      "grad_norm": 0.11631784496079337,
      "learning_rate": 9.904626120179505e-07,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001590099884197116,
      "step": 1530,
      "valid_targets_mean": 1350.9,
      "valid_targets_min": 590
    },
    {
      "epoch": 7.525798525798526,
      "grad_norm": 0.12342575829593498,
      "learning_rate": 8.969167909475939e-07,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002005585702136159,
      "step": 1535,
      "valid_targets_mean": 1383.1,
      "valid_targets_min": 891
    },
    {
      "epoch": 7.5503685503685505,
      "grad_norm": 0.1377112054055627,
      "learning_rate": 8.079704430163204e-07,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001176726073026657,
      "step": 1540,
      "valid_targets_mean": 1381.1,
      "valid_targets_min": 819
    },
    {
      "epoch": 7.5749385749385745,
      "grad_norm": 0.12267343278939448,
      "learning_rate": 7.23631899805588e-07,
      "loss": 0.0026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001982441172003746,
      "step": 1545,
      "valid_targets_mean": 1264.6,
      "valid_targets_min": 563
    },
    {
      "epoch": 7.599508599508599,
      "grad_norm": 0.1232331023445151,
      "learning_rate": 6.439090612849863e-07,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0012900140136480331,
      "step": 1550,
      "valid_targets_mean": 1392.8,
      "valid_targets_min": 744
    },
    {
      "epoch": 7.624078624078624,
      "grad_norm": 0.12399612774127011,
      "learning_rate": 5.688093950722451e-07,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0012250710278749466,
      "step": 1555,
      "valid_targets_mean": 1380.9,
      "valid_targets_min": 823
    },
    {
      "epoch": 7.648648648648649,
      "grad_norm": 0.10093776597270249,
      "learning_rate": 4.983399357337215e-07,
      "loss": 0.0024,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0011251834221184254,
      "step": 1560,
      "valid_targets_mean": 1338.3,
      "valid_targets_min": 745
    },
    {
      "epoch": 7.673218673218673,
      "grad_norm": 0.14023903435720927,
      "learning_rate": 4.32507284125494e-07,
      "loss": 0.0026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0011913267662748694,
      "step": 1565,
      "valid_targets_mean": 1174.8,
      "valid_targets_min": 654
    },
    {
      "epoch": 7.697788697788698,
      "grad_norm": 0.11347502220023985,
      "learning_rate": 3.7131760677505676e-07,
      "loss": 0.0023,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0011188078206032515,
      "step": 1570,
      "valid_targets_mean": 1516.1,
      "valid_targets_min": 749
    },
    {
      "epoch": 7.722358722358722,
      "grad_norm": 0.09825121141173124,
      "learning_rate": 3.1477663530371514e-07,
      "loss": 0.0021,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001625724951736629,
      "step": 1575,
      "valid_targets_mean": 1451.0,
      "valid_targets_min": 828
    },
    {
      "epoch": 7.746928746928747,
      "grad_norm": 0.15120852978627652,
      "learning_rate": 2.6288966588967623e-07,
      "loss": 0.0026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0013657582458108664,
      "step": 1580,
      "valid_targets_mean": 1270.2,
      "valid_targets_min": 615
    },
    {
      "epoch": 7.771498771498772,
      "grad_norm": 0.09585305972292815,
      "learning_rate": 2.1566155877197903e-07,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0007987943245097995,
      "step": 1585,
      "valid_targets_mean": 1350.7,
      "valid_targets_min": 609
    },
    {
      "epoch": 7.796068796068796,
      "grad_norm": 0.13596142729213964,
      "learning_rate": 1.7309673779524194e-07,
      "loss": 0.0023,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001020238851197064,
      "step": 1590,
      "valid_targets_mean": 1383.6,
      "valid_targets_min": 812
    },
    {
      "epoch": 7.820638820638821,
      "grad_norm": 0.10746814374827729,
      "learning_rate": 1.3519918999526648e-07,
      "loss": 0.0023,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0010407338850200176,
      "step": 1595,
      "valid_targets_mean": 1373.1,
      "valid_targets_min": 605
    },
    {
      "epoch": 7.8452088452088455,
      "grad_norm": 0.1366405061352433,
      "learning_rate": 1.0197246522557491e-07,
      "loss": 0.0026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001199243706651032,
      "step": 1600,
      "valid_targets_mean": 1318.0,
      "valid_targets_min": 619
    },
    {
      "epoch": 7.8697788697788695,
      "grad_norm": 0.15509314338643704,
      "learning_rate": 7.34196758249095e-08,
      "loss": 0.0038,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0018726876005530357,
      "step": 1605,
      "valid_targets_mean": 1289.9,
      "valid_targets_min": 800
    },
    {
      "epoch": 7.894348894348894,
      "grad_norm": 0.11868071753447657,
      "learning_rate": 4.9543496325693553e-08,
      "loss": 0.0022,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0009430464124307036,
      "step": 1610,
      "valid_targets_mean": 1305.1,
      "valid_targets_min": 719
    },
    {
      "epoch": 7.918918918918919,
      "grad_norm": 0.10845984516950594,
      "learning_rate": 3.034616320349293e-08,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0009824966546148062,
      "step": 1615,
      "valid_targets_mean": 1196.1,
      "valid_targets_min": 518
    },
    {
      "epoch": 7.943488943488943,
      "grad_norm": 0.1252822512833604,
      "learning_rate": 1.58294746675558e-08,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0012963269837200642,
      "step": 1620,
      "valid_targets_mean": 1239.3,
      "valid_targets_min": 587
    },
    {
      "epoch": 7.968058968058968,
      "grad_norm": 0.113750323811112,
      "learning_rate": 5.994790492352964e-09,
      "loss": 0.0025,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0010778913274407387,
      "step": 1625,
      "valid_targets_mean": 1516.6,
      "valid_targets_min": 668
    },
    {
      "epoch": 7.992628992628992,
      "grad_norm": 0.11051935412163597,
      "learning_rate": 8.43031890213064e-10,
      "loss": 0.0024,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.001104144612327218,
      "step": 1630,
      "valid_targets_mean": 1378.2,
      "valid_targets_min": 825
    },
    {
      "epoch": 8.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0019607325084507465,
      "step": 1632,
      "total_flos": 3.419079710112481e+17,
      "train_loss": 0.08937609711645957,
      "train_runtime": 8905.9814,
      "train_samples_per_second": 5.841,
      "train_steps_per_second": 0.183,
      "valid_targets_mean": 1431.2,
      "valid_targets_min": 619
    }
  ],
  "logging_steps": 5,
  "max_steps": 1632,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.419079710112481e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}