{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 785,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032,
      "grad_norm": 10.231114400595384,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 1.0228,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.26492345333099365,
      "step": 5,
      "valid_targets_mean": 1423.7,
      "valid_targets_min": 820
    },
    {
      "epoch": 0.064,
      "grad_norm": 7.039277324325124,
      "learning_rate": 4.556962025316456e-06,
      "loss": 1.0089,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.24275313317775726,
      "step": 10,
      "valid_targets_mean": 1370.4,
      "valid_targets_min": 842
    },
    {
      "epoch": 0.096,
      "grad_norm": 3.1172704392402246,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.9329,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.2439592480659485,
      "step": 15,
      "valid_targets_mean": 1500.3,
      "valid_targets_min": 980
    },
    {
      "epoch": 0.128,
      "grad_norm": 1.6954429191729048,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.8713,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.22679400444030762,
      "step": 20,
      "valid_targets_mean": 1494.3,
      "valid_targets_min": 834
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.263351838373193,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.8061,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.19422025978565216,
      "step": 25,
      "valid_targets_mean": 1280.0,
      "valid_targets_min": 627
    },
    {
      "epoch": 0.192,
      "grad_norm": 1.3088208699740986,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.7795,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1766110062599182,
      "step": 30,
      "valid_targets_mean": 1336.8,
      "valid_targets_min": 695
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.7746922535731153,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.7298,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1887187361717224,
      "step": 35,
      "valid_targets_mean": 1429.8,
      "valid_targets_min": 746
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.6246271827449218,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.711,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17022612690925598,
      "step": 40,
      "valid_targets_mean": 1697.3,
      "valid_targets_min": 765
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.5504111442203626,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.662,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14026479423046112,
      "step": 45,
      "valid_targets_mean": 1342.4,
      "valid_targets_min": 505
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5316671576883836,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.6548,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1596953421831131,
      "step": 50,
      "valid_targets_mean": 1421.1,
      "valid_targets_min": 891
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.5097700775340454,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.6228,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14719301462173462,
      "step": 55,
      "valid_targets_mean": 1353.8,
      "valid_targets_min": 696
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.48145906073164757,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.6103,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1529427170753479,
      "step": 60,
      "valid_targets_mean": 1494.7,
      "valid_targets_min": 748
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.5146437130801124,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.6253,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1409555822610855,
      "step": 65,
      "valid_targets_mean": 1246.2,
      "valid_targets_min": 666
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.509891659300274,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.5946,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1452096402645111,
      "step": 70,
      "valid_targets_mean": 1405.4,
      "valid_targets_min": 770
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5184798975735082,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.6023,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1414271742105484,
      "step": 75,
      "valid_targets_mean": 1471.4,
      "valid_targets_min": 442
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.5185493191792633,
      "learning_rate": 4e-05,
      "loss": 0.5754,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1363818347454071,
      "step": 80,
      "valid_targets_mean": 1258.1,
      "valid_targets_min": 706
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.48733095905238816,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.5726,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13037832081317902,
      "step": 85,
      "valid_targets_mean": 1328.4,
      "valid_targets_min": 659
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.4781888097489448,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.5743,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13059714436531067,
      "step": 90,
      "valid_targets_mean": 1344.5,
      "valid_targets_min": 670
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.5138407373391729,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.566,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14600546658039093,
      "step": 95,
      "valid_targets_mean": 1510.1,
      "valid_targets_min": 879
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.48614664202485364,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.556,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14747026562690735,
      "step": 100,
      "valid_targets_mean": 1437.4,
      "valid_targets_min": 886
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.4947657903827249,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.5561,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15439683198928833,
      "step": 105,
      "valid_targets_mean": 1617.1,
      "valid_targets_min": 760
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.47018544010663216,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.5448,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1327132284641266,
      "step": 110,
      "valid_targets_mean": 1411.9,
      "valid_targets_min": 890
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.4846145638177327,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.5526,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.139687642455101,
      "step": 115,
      "valid_targets_mean": 1305.5,
      "valid_targets_min": 576
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.5064084410597361,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.5474,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1417393684387207,
      "step": 120,
      "valid_targets_mean": 1442.5,
      "valid_targets_min": 685
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.4939280731102979,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.5495,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13436730206012726,
      "step": 125,
      "valid_targets_mean": 1470.2,
      "valid_targets_min": 603
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.48423589358193253,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.5538,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14917421340942383,
      "step": 130,
      "valid_targets_mean": 1358.9,
      "valid_targets_min": 687
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.47290787462943296,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.5436,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1280403435230255,
      "step": 135,
      "valid_targets_mean": 1350.4,
      "valid_targets_min": 496
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.4944509656442681,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.5232,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1382235586643219,
      "step": 140,
      "valid_targets_mean": 1361.7,
      "valid_targets_min": 729
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.4897245405773179,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.5291,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14631761610507965,
      "step": 145,
      "valid_targets_mean": 1458.4,
      "valid_targets_min": 812
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.47533001709887873,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.5323,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1408248245716095,
      "step": 150,
      "valid_targets_mean": 1502.6,
      "valid_targets_min": 674
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.4925753589851704,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.5166,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12974439561367035,
      "step": 155,
      "valid_targets_mean": 1537.8,
      "valid_targets_min": 676
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.4992903485783338,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.5329,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12389901280403137,
      "step": 160,
      "valid_targets_mean": 1364.3,
      "valid_targets_min": 575
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.5198494923150886,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.526,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13451017439365387,
      "step": 165,
      "valid_targets_mean": 1301.1,
      "valid_targets_min": 683
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.4933804252349336,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.5174,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13121582567691803,
      "step": 170,
      "valid_targets_mean": 1527.9,
      "valid_targets_min": 858
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.5261399898137121,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.5099,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12689976394176483,
      "step": 175,
      "valid_targets_mean": 1376.9,
      "valid_targets_min": 535
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.46777081663807835,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.5037,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10239965468645096,
      "step": 180,
      "valid_targets_mean": 1281.8,
      "valid_targets_min": 932
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.4676054248162362,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.4953,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1201857402920723,
      "step": 185,
      "valid_targets_mean": 1340.0,
      "valid_targets_min": 826
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.47557823247680286,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.5085,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11424747109413147,
      "step": 190,
      "valid_targets_mean": 1379.6,
      "valid_targets_min": 735
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.4474396748536332,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.5052,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12652286887168884,
      "step": 195,
      "valid_targets_mean": 1732.6,
      "valid_targets_min": 1083
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.5071822384645137,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.5089,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11714669317007065,
      "step": 200,
      "valid_targets_mean": 1269.4,
      "valid_targets_min": 824
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.4945774512412606,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.5087,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12087100744247437,
      "step": 205,
      "valid_targets_mean": 1436.1,
      "valid_targets_min": 712
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.4816631088932476,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.5205,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14656808972358704,
      "step": 210,
      "valid_targets_mean": 1706.1,
      "valid_targets_min": 943
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.4959091389971953,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.5131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12078063189983368,
      "step": 215,
      "valid_targets_mean": 1234.4,
      "valid_targets_min": 544
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.4594939627264897,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.5017,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12703871726989746,
      "step": 220,
      "valid_targets_mean": 1642.6,
      "valid_targets_min": 615
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.47136825772108737,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.5079,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1292632818222046,
      "step": 225,
      "valid_targets_mean": 1385.5,
      "valid_targets_min": 747
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.4659865429362735,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.5026,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14096719026565552,
      "step": 230,
      "valid_targets_mean": 1798.4,
      "valid_targets_min": 1079
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.4880684467688744,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.5013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13877353072166443,
      "step": 235,
      "valid_targets_mean": 1640.6,
      "valid_targets_min": 891
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.47056177557765516,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.5066,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10828080773353577,
      "step": 240,
      "valid_targets_mean": 1284.7,
      "valid_targets_min": 627
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.4778498930456995,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.5124,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11812753975391388,
      "step": 245,
      "valid_targets_mean": 1256.3,
      "valid_targets_min": 862
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.5172655499323758,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.5083,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11932039260864258,
      "step": 250,
      "valid_targets_mean": 1254.3,
      "valid_targets_min": 918
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.5025983697311104,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.5022,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11708801984786987,
      "step": 255,
      "valid_targets_mean": 1424.1,
      "valid_targets_min": 699
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.47271725241360835,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.5029,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11993420124053955,
      "step": 260,
      "valid_targets_mean": 1424.9,
      "valid_targets_min": 815
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.4775633734349197,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.5047,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12442141771316528,
      "step": 265,
      "valid_targets_mean": 1310.2,
      "valid_targets_min": 616
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.47543978446782803,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.5005,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13535171747207642,
      "step": 270,
      "valid_targets_mean": 1439.1,
      "valid_targets_min": 834
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.4821229703460132,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.503,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13594815135002136,
      "step": 275,
      "valid_targets_mean": 1513.6,
      "valid_targets_min": 720
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.47694704715775016,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.5004,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13347353041172028,
      "step": 280,
      "valid_targets_mean": 1436.6,
      "valid_targets_min": 739
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.46948934236738865,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.4903,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1326732039451599,
      "step": 285,
      "valid_targets_mean": 1448.5,
      "valid_targets_min": 994
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.49081890650616944,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.4965,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11286979913711548,
      "step": 290,
      "valid_targets_mean": 1451.4,
      "valid_targets_min": 672
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.4733499931855284,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.5105,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13379234075546265,
      "step": 295,
      "valid_targets_mean": 1246.1,
      "valid_targets_min": 805
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.48654201912855316,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.4953,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14153556525707245,
      "step": 300,
      "valid_targets_mean": 1494.7,
      "valid_targets_min": 718
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.5070760726209889,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.4888,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12899258732795715,
      "step": 305,
      "valid_targets_mean": 1468.2,
      "valid_targets_min": 605
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.49244904749996854,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.4986,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11913871765136719,
      "step": 310,
      "valid_targets_mean": 1252.3,
      "valid_targets_min": 573
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.5212910931679563,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.489,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14036613702774048,
      "step": 315,
      "valid_targets_mean": 1364.4,
      "valid_targets_min": 668
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.5151332262167365,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.4747,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12699592113494873,
      "step": 320,
      "valid_targets_mean": 1364.8,
      "valid_targets_min": 842
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.4572624287771293,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.4795,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12322171032428741,
      "step": 325,
      "valid_targets_mean": 1588.8,
      "valid_targets_min": 962
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.5039021854464919,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.4761,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1055314838886261,
      "step": 330,
      "valid_targets_mean": 1334.5,
      "valid_targets_min": 631
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.49100221715857273,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.4762,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12410566210746765,
      "step": 335,
      "valid_targets_mean": 1504.7,
      "valid_targets_min": 831
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.5728430459921913,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.4892,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1243298128247261,
      "step": 340,
      "valid_targets_mean": 1304.8,
      "valid_targets_min": 798
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.4869178118536037,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.4745,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1093498021364212,
      "step": 345,
      "valid_targets_mean": 1452.5,
      "valid_targets_min": 1018
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.442288500421061,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.4717,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10126109421253204,
      "step": 350,
      "valid_targets_mean": 1409.2,
      "valid_targets_min": 773
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.47194674591478103,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.4712,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11688213795423508,
      "step": 355,
      "valid_targets_mean": 1433.2,
      "valid_targets_min": 779
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.48432150731864476,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.4702,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12231733649969101,
      "step": 360,
      "valid_targets_mean": 1484.7,
      "valid_targets_min": 720
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.4577949671383314,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.4721,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11381445825099945,
      "step": 365,
      "valid_targets_mean": 1448.4,
      "valid_targets_min": 896
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.44337673811838285,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.4676,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10979792475700378,
      "step": 370,
      "valid_targets_mean": 1318.5,
      "valid_targets_min": 734
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.49537569548309385,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.4796,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11864444613456726,
      "step": 375,
      "valid_targets_mean": 1452.1,
      "valid_targets_min": 475
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.5315521183598081,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.482,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12169407308101654,
      "step": 380,
      "valid_targets_mean": 1419.9,
      "valid_targets_min": 724
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.49506460908083316,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.4718,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12358853220939636,
      "step": 385,
      "valid_targets_mean": 1520.4,
      "valid_targets_min": 908
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.48761547554498685,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.4832,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11058260500431061,
      "step": 390,
      "valid_targets_mean": 1306.4,
      "valid_targets_min": 603
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.46175521448983564,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.4831,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11915287375450134,
      "step": 395,
      "valid_targets_mean": 1440.4,
      "valid_targets_min": 1134
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.47265880179829556,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.4763,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11093438416719437,
      "step": 400,
      "valid_targets_mean": 1310.6,
      "valid_targets_min": 484
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.47847619166876576,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.4689,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1183091402053833,
      "step": 405,
      "valid_targets_mean": 1459.4,
      "valid_targets_min": 589
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.47213163716287915,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.4856,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12554430961608887,
      "step": 410,
      "valid_targets_mean": 1516.8,
      "valid_targets_min": 573
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.4318914817798841,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.4696,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11150006204843521,
      "step": 415,
      "valid_targets_mean": 1610.6,
      "valid_targets_min": 615
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.4654113860974772,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.4706,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11771731078624725,
      "step": 420,
      "valid_targets_mean": 1449.2,
      "valid_targets_min": 644
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.4495611716753332,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.4726,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12123566120862961,
      "step": 425,
      "valid_targets_mean": 1672.8,
      "valid_targets_min": 920
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.517347703585648,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.477,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1257917881011963,
      "step": 430,
      "valid_targets_mean": 1405.6,
      "valid_targets_min": 585
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.47476629971205914,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.4749,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12210121750831604,
      "step": 435,
      "valid_targets_mean": 1393.9,
      "valid_targets_min": 570
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.42585526362825243,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.459,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12968218326568604,
      "step": 440,
      "valid_targets_mean": 1921.7,
      "valid_targets_min": 966
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.4731019812725222,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.4741,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11867497861385345,
      "step": 445,
      "valid_targets_mean": 1509.7,
      "valid_targets_min": 972
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.4648950793860925,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.4697,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11365693062543869,
      "step": 450,
      "valid_targets_mean": 1382.8,
      "valid_targets_min": 535
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.4881561152181817,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.4744,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12787684798240662,
      "step": 455,
      "valid_targets_mean": 1308.9,
      "valid_targets_min": 667
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.48326188754148974,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.4867,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12152108550071716,
      "step": 460,
      "valid_targets_mean": 1302.2,
      "valid_targets_min": 494
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.47050962159189763,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.4712,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11341217160224915,
      "step": 465,
      "valid_targets_mean": 1364.2,
      "valid_targets_min": 783
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.507355236705902,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.4759,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14260517060756683,
      "step": 470,
      "valid_targets_mean": 1551.1,
      "valid_targets_min": 766
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.43919926941057863,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.4557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11649820953607559,
      "step": 475,
      "valid_targets_mean": 1684.8,
      "valid_targets_min": 973
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.5113336451827102,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.4573,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12410803139209747,
      "step": 480,
      "valid_targets_mean": 1396.0,
      "valid_targets_min": 597
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.45571766591709284,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.4579,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11411958932876587,
      "step": 485,
      "valid_targets_mean": 1413.1,
      "valid_targets_min": 809
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.48463017803010616,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.457,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10860386490821838,
      "step": 490,
      "valid_targets_mean": 1355.2,
      "valid_targets_min": 560
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.5352572529710553,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.4564,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11461152136325836,
      "step": 495,
      "valid_targets_mean": 1542.4,
      "valid_targets_min": 861
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.4617604351088588,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.4619,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11053967475891113,
      "step": 500,
      "valid_targets_mean": 1437.3,
      "valid_targets_min": 761
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.4631766233510783,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.4648,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11951716989278793,
      "step": 505,
      "valid_targets_mean": 1500.6,
      "valid_targets_min": 610
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.4931434969673906,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.4537,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09979387372732162,
      "step": 510,
      "valid_targets_mean": 1245.6,
      "valid_targets_min": 580
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.47110754618641665,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.463,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11337228119373322,
      "step": 515,
      "valid_targets_mean": 1454.6,
      "valid_targets_min": 724
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.46227921260170174,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.4559,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12954063713550568,
      "step": 520,
      "valid_targets_mean": 1497.1,
      "valid_targets_min": 1160
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.4663726713269175,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.4499,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10027768462896347,
      "step": 525,
      "valid_targets_mean": 1202.9,
      "valid_targets_min": 582
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.5206370393525042,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.4478,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11333752423524857,
      "step": 530,
      "valid_targets_mean": 1612.1,
      "valid_targets_min": 607
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.47775195380747376,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.4663,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12556275725364685,
      "step": 535,
      "valid_targets_mean": 1522.3,
      "valid_targets_min": 760
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.4593481544070819,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.4709,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11827541887760162,
      "step": 540,
      "valid_targets_mean": 1497.2,
      "valid_targets_min": 855
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.47156511303943216,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.454,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10690443217754364,
      "step": 545,
      "valid_targets_mean": 1238.9,
      "valid_targets_min": 592
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.4766796375881957,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.4667,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11884674429893494,
      "step": 550,
      "valid_targets_mean": 1447.5,
      "valid_targets_min": 870
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.466356604354509,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.4663,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11681554466485977,
      "step": 555,
      "valid_targets_mean": 1318.4,
      "valid_targets_min": 775
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.45690110419565866,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.4604,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11520233750343323,
      "step": 560,
      "valid_targets_mean": 1461.6,
      "valid_targets_min": 544
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.4675602081663667,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.4616,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09840774536132812,
      "step": 565,
      "valid_targets_mean": 1351.1,
      "valid_targets_min": 762
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.4553233405365835,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.4538,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10833073407411575,
      "step": 570,
      "valid_targets_mean": 1405.3,
      "valid_targets_min": 627
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.47193088669734595,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.4597,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12648598849773407,
      "step": 575,
      "valid_targets_mean": 1574.1,
      "valid_targets_min": 934
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.4429748408387231,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.4578,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1061725914478302,
      "step": 580,
      "valid_targets_mean": 1469.0,
      "valid_targets_min": 818
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.4898833206908071,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.4642,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12915876507759094,
      "step": 585,
      "valid_targets_mean": 1447.4,
      "valid_targets_min": 748
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.42553530049589794,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.4567,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10126683861017227,
      "step": 590,
      "valid_targets_mean": 1355.8,
      "valid_targets_min": 725
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.4317363804249038,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.4514,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10533666610717773,
      "step": 595,
      "valid_targets_mean": 1314.1,
      "valid_targets_min": 837
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.4428922495764163,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.458,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11306220293045044,
      "step": 600,
      "valid_targets_mean": 1486.4,
      "valid_targets_min": 673
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.45915453646201937,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.4557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10991701483726501,
      "step": 605,
      "valid_targets_mean": 1211.1,
      "valid_targets_min": 672
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.5124726416089546,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.4637,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11427893489599228,
      "step": 610,
      "valid_targets_mean": 1308.8,
      "valid_targets_min": 633
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.4595443969121929,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.4578,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1211710274219513,
      "step": 615,
      "valid_targets_mean": 1570.5,
      "valid_targets_min": 841
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.4838441082754233,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.4616,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1310589760541916,
      "step": 620,
      "valid_targets_mean": 1488.1,
      "valid_targets_min": 891
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.4463739198336122,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.4514,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10531221330165863,
      "step": 625,
      "valid_targets_mean": 1620.9,
      "valid_targets_min": 994
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.47465817729128684,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.4413,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10876576602458954,
      "step": 630,
      "valid_targets_mean": 1388.1,
      "valid_targets_min": 779
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.47666648434482584,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.4546,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11096735298633575,
      "step": 635,
      "valid_targets_mean": 1179.4,
      "valid_targets_min": 610
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.446761899374689,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.4434,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12088684737682343,
      "step": 640,
      "valid_targets_mean": 1404.6,
      "valid_targets_min": 1075
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.45442678304961115,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.4557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11140430718660355,
      "step": 645,
      "valid_targets_mean": 1330.2,
      "valid_targets_min": 679
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.44594175759108007,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.4497,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10718557238578796,
      "step": 650,
      "valid_targets_mean": 1326.6,
      "valid_targets_min": 524
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.4411885157805194,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.4363,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11127768456935883,
      "step": 655,
      "valid_targets_mean": 1552.0,
      "valid_targets_min": 1046
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.4790427115520168,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.4558,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12272059172391891,
      "step": 660,
      "valid_targets_mean": 1381.2,
      "valid_targets_min": 900
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.4609725571539068,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.4528,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10866587609052658,
      "step": 665,
      "valid_targets_mean": 1343.3,
      "valid_targets_min": 689
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.45993621820043235,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.4505,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0955156683921814,
      "step": 670,
      "valid_targets_mean": 1296.0,
      "valid_targets_min": 633
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.49225870110490405,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.4357,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11676684767007828,
      "step": 675,
      "valid_targets_mean": 1561.1,
      "valid_targets_min": 814
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.4431187464482393,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.453,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11237336695194244,
      "step": 680,
      "valid_targets_mean": 1550.1,
      "valid_targets_min": 667
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.45933291980777213,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.4596,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11005580425262451,
      "step": 685,
      "valid_targets_mean": 1332.4,
      "valid_targets_min": 810
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.4281088914703306,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.4493,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1104857474565506,
      "step": 690,
      "valid_targets_mean": 1489.4,
      "valid_targets_min": 562
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.4499426572561623,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.4495,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11642653495073318,
      "step": 695,
      "valid_targets_mean": 1511.1,
      "valid_targets_min": 697
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.49654084278921,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.4544,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10965511202812195,
      "step": 700,
      "valid_targets_mean": 1283.9,
      "valid_targets_min": 693
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.4410552186345025,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.4491,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11442539840936661,
      "step": 705,
      "valid_targets_mean": 1500.0,
      "valid_targets_min": 989
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.437211441585052,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.4461,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12135186791419983,
      "step": 710,
      "valid_targets_mean": 1468.4,
      "valid_targets_min": 832
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.44268840936929776,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.4506,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12368033826351166,
      "step": 715,
      "valid_targets_mean": 1538.1,
      "valid_targets_min": 960
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.4417605859691051,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.4541,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11311336606740952,
      "step": 720,
      "valid_targets_mean": 1383.7,
      "valid_targets_min": 904
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.46572331077704227,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.455,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13641811907291412,
      "step": 725,
      "valid_targets_mean": 1593.6,
      "valid_targets_min": 988
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.4496105188871889,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.4518,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10604508221149445,
      "step": 730,
      "valid_targets_mean": 1385.3,
      "valid_targets_min": 627
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.4745802069109058,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.4554,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12264803797006607,
      "step": 735,
      "valid_targets_mean": 1405.0,
      "valid_targets_min": 683
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.4670645740261794,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.4472,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10746762156486511,
      "step": 740,
      "valid_targets_mean": 1299.8,
      "valid_targets_min": 719
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.4555586350543962,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.4476,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10571619868278503,
      "step": 745,
      "valid_targets_mean": 1316.4,
      "valid_targets_min": 815
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.4615252786013569,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.4486,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11292578279972076,
      "step": 750,
      "valid_targets_mean": 1289.4,
      "valid_targets_min": 747
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.41887534788483755,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.4486,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11504572629928589,
      "step": 755,
      "valid_targets_mean": 1803.6,
      "valid_targets_min": 580
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.45162607038670965,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.4574,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09912922978401184,
      "step": 760,
      "valid_targets_mean": 1297.7,
      "valid_targets_min": 762
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.4233976282554638,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.4484,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12133409827947617,
      "step": 765,
      "valid_targets_mean": 1630.6,
      "valid_targets_min": 693
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.45514909271692383,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.4449,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10166063159704208,
      "step": 770,
      "valid_targets_mean": 1396.9,
      "valid_targets_min": 668
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.4516233378576753,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.449,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11956135183572769,
      "step": 775,
      "valid_targets_mean": 1427.9,
      "valid_targets_min": 749
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.44561728138524365,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.4587,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11012592166662216,
      "step": 780,
      "valid_targets_mean": 1292.1,
      "valid_targets_min": 580
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.3093916982081615,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.4528,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4559069871902466,
      "step": 785,
      "valid_targets_mean": 1217.6,
      "valid_targets_min": 645
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4559069871902466,
      "step": 785,
      "total_flos": 4.2611210572110234e+17,
      "train_loss": 0.5067302330284361,
      "train_runtime": 6941.844,
      "train_samples_per_second": 7.201,
      "train_steps_per_second": 0.113,
      "valid_targets_mean": 1217.6,
      "valid_targets_min": 645
    }
  ],
  "logging_steps": 5,
  "max_steps": 785,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.2611210572110234e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}