{
  "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": 13.448105258759517,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 1.2075,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.31800004839897156,
      "step": 5,
      "valid_targets_mean": 748.6,
      "valid_targets_min": 403
    },
    {
      "epoch": 0.064,
      "grad_norm": 9.982424737572133,
      "learning_rate": 4.556962025316456e-06,
      "loss": 1.1769,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.3006700873374939,
      "step": 10,
      "valid_targets_mean": 714.3,
      "valid_targets_min": 468
    },
    {
      "epoch": 0.096,
      "grad_norm": 3.8464182382921748,
      "learning_rate": 7.08860759493671e-06,
      "loss": 1.0706,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.2761875092983246,
      "step": 15,
      "valid_targets_mean": 833.6,
      "valid_targets_min": 428
    },
    {
      "epoch": 0.128,
      "grad_norm": 2.1620562822697003,
      "learning_rate": 9.620253164556963e-06,
      "loss": 1.0013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.25137242674827576,
      "step": 20,
      "valid_targets_mean": 814.9,
      "valid_targets_min": 392
    },
    {
      "epoch": 0.16,
      "grad_norm": 1.6337626383709007,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.9265,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.2275439351797104,
      "step": 25,
      "valid_targets_mean": 728.6,
      "valid_targets_min": 416
    },
    {
      "epoch": 0.192,
      "grad_norm": 1.9026845295535284,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.8832,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.20272783935070038,
      "step": 30,
      "valid_targets_mean": 751.1,
      "valid_targets_min": 439
    },
    {
      "epoch": 0.224,
      "grad_norm": 1.073091888557403,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.8224,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.21370205283164978,
      "step": 35,
      "valid_targets_mean": 786.2,
      "valid_targets_min": 463
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.8617650331409201,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.8024,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1904214322566986,
      "step": 40,
      "valid_targets_mean": 843.4,
      "valid_targets_min": 359
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.759980239581234,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.7507,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15183305740356445,
      "step": 45,
      "valid_targets_mean": 724.6,
      "valid_targets_min": 343
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.7463412043901649,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.736,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1823093295097351,
      "step": 50,
      "valid_targets_mean": 826.6,
      "valid_targets_min": 562
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.6886141045868305,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.7099,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17409449815750122,
      "step": 55,
      "valid_targets_mean": 738.2,
      "valid_targets_min": 387
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.6723991947723206,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.6817,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1746065616607666,
      "step": 60,
      "valid_targets_mean": 817.4,
      "valid_targets_min": 503
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.7219762987855783,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.7042,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15542849898338318,
      "step": 65,
      "valid_targets_mean": 631.1,
      "valid_targets_min": 418
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.711945449946034,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.6683,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1588052213191986,
      "step": 70,
      "valid_targets_mean": 717.1,
      "valid_targets_min": 483
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.7723131517405141,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.6602,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16386288404464722,
      "step": 75,
      "valid_targets_mean": 850.6,
      "valid_targets_min": 252
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.7455251121043946,
      "learning_rate": 4e-05,
      "loss": 0.6492,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16289247572422028,
      "step": 80,
      "valid_targets_mean": 734.2,
      "valid_targets_min": 385
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.6806696789899241,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.6491,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14095750451087952,
      "step": 85,
      "valid_targets_mean": 764.7,
      "valid_targets_min": 487
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.6997408572409954,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.6471,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15331107378005981,
      "step": 90,
      "valid_targets_mean": 785.2,
      "valid_targets_min": 492
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.7180343376382498,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.6481,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17766906321048737,
      "step": 95,
      "valid_targets_mean": 781.8,
      "valid_targets_min": 397
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.7022835185401193,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.6207,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.17354968190193176,
      "step": 100,
      "valid_targets_mean": 836.8,
      "valid_targets_min": 418
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.7367281503709805,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.6272,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.162840336561203,
      "step": 105,
      "valid_targets_mean": 837.1,
      "valid_targets_min": 434
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.6596850958225633,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.6176,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14747610688209534,
      "step": 110,
      "valid_targets_mean": 715.6,
      "valid_targets_min": 485
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.6740339776230146,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.6183,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15198436379432678,
      "step": 115,
      "valid_targets_mean": 681.2,
      "valid_targets_min": 400
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.6279891719181242,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.6138,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1494234800338745,
      "step": 120,
      "valid_targets_mean": 762.9,
      "valid_targets_min": 313
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.6762014593224054,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.6152,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14409850537776947,
      "step": 125,
      "valid_targets_mean": 826.2,
      "valid_targets_min": 432
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.6951704276499515,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.6211,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1555303931236267,
      "step": 130,
      "valid_targets_mean": 755.7,
      "valid_targets_min": 496
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.6296778796453734,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.6129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1367054283618927,
      "step": 135,
      "valid_targets_mean": 662.6,
      "valid_targets_min": 328
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.6783518650883749,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.5886,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15633609890937805,
      "step": 140,
      "valid_targets_mean": 727.8,
      "valid_targets_min": 507
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.708146207975277,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.5948,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16752688586711884,
      "step": 145,
      "valid_targets_mean": 762.7,
      "valid_targets_min": 541
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.6593931091646447,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.5933,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.164844810962677,
      "step": 150,
      "valid_targets_mean": 851.9,
      "valid_targets_min": 389
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.6510250678307847,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.5897,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1499878615140915,
      "step": 155,
      "valid_targets_mean": 824.8,
      "valid_targets_min": 424
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.6781043016492474,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.5918,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1404564082622528,
      "step": 160,
      "valid_targets_mean": 745.8,
      "valid_targets_min": 381
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.6508626538105656,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.5805,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14573101699352264,
      "step": 165,
      "valid_targets_mean": 690.4,
      "valid_targets_min": 426
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.6313686061473457,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.5801,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13861840963363647,
      "step": 170,
      "valid_targets_mean": 763.2,
      "valid_targets_min": 457
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.6718637932023319,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.5727,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13479197025299072,
      "step": 175,
      "valid_targets_mean": 740.8,
      "valid_targets_min": 325
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.6534534284351228,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.5688,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11501454561948776,
      "step": 180,
      "valid_targets_mean": 669.1,
      "valid_targets_min": 361
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.6298087553175363,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.5533,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12853585183620453,
      "step": 185,
      "valid_targets_mean": 759.6,
      "valid_targets_min": 433
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.6629648321914637,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.5718,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13259121775627136,
      "step": 190,
      "valid_targets_mean": 673.8,
      "valid_targets_min": 428
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.6446910197966146,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.5721,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1383952796459198,
      "step": 195,
      "valid_targets_mean": 777.4,
      "valid_targets_min": 559
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.6715593999996728,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.5767,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13751399517059326,
      "step": 200,
      "valid_targets_mean": 690.3,
      "valid_targets_min": 362
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.6634970200747077,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.5749,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14345307648181915,
      "step": 205,
      "valid_targets_mean": 795.1,
      "valid_targets_min": 436
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.6852785997622871,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.5757,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16555346548557281,
      "step": 210,
      "valid_targets_mean": 953.1,
      "valid_targets_min": 495
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.7182921119480419,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.5662,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13560381531715393,
      "step": 215,
      "valid_targets_mean": 657.9,
      "valid_targets_min": 380
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.6383915774320499,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.5598,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1385164111852646,
      "step": 220,
      "valid_targets_mean": 883.5,
      "valid_targets_min": 351
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.6197700397838696,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.5645,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12843455374240875,
      "step": 225,
      "valid_targets_mean": 750.9,
      "valid_targets_min": 434
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.5711420210151498,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.5598,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14530006051063538,
      "step": 230,
      "valid_targets_mean": 942.8,
      "valid_targets_min": 528
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.6210204975257504,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.5593,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1555243879556656,
      "step": 235,
      "valid_targets_mean": 887.2,
      "valid_targets_min": 537
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.6647791326711432,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.5609,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13146094977855682,
      "step": 240,
      "valid_targets_mean": 744.5,
      "valid_targets_min": 461
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.6693133474516781,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.577,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1364261358976364,
      "step": 245,
      "valid_targets_mean": 682.8,
      "valid_targets_min": 444
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.7006371063676372,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.5673,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13232733309268951,
      "step": 250,
      "valid_targets_mean": 724.2,
      "valid_targets_min": 468
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.655773431685838,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.561,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13282510638237,
      "step": 255,
      "valid_targets_mean": 766.9,
      "valid_targets_min": 402
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.6172192241649032,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.5643,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13580721616744995,
      "step": 260,
      "valid_targets_mean": 764.2,
      "valid_targets_min": 400
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.6628955705415062,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.5581,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1439439356327057,
      "step": 265,
      "valid_targets_mean": 715.3,
      "valid_targets_min": 309
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.641312579448422,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.5624,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1477782130241394,
      "step": 270,
      "valid_targets_mean": 739.2,
      "valid_targets_min": 472
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.6833918688567646,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.5578,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15322580933570862,
      "step": 275,
      "valid_targets_mean": 752.0,
      "valid_targets_min": 463
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.640918224676768,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.5599,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13823790848255157,
      "step": 280,
      "valid_targets_mean": 789.4,
      "valid_targets_min": 321
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.6654266115883691,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.5485,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14035166800022125,
      "step": 285,
      "valid_targets_mean": 807.4,
      "valid_targets_min": 448
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.6123014675225724,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.5557,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12820711731910706,
      "step": 290,
      "valid_targets_mean": 850.4,
      "valid_targets_min": 424
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.6754595695415974,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.5627,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15113355219364166,
      "step": 295,
      "valid_targets_mean": 696.0,
      "valid_targets_min": 472
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.6578705510880959,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.5519,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15135137736797333,
      "step": 300,
      "valid_targets_mean": 767.6,
      "valid_targets_min": 517
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.6449455299555445,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.548,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14119261503219604,
      "step": 305,
      "valid_targets_mean": 762.3,
      "valid_targets_min": 434
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.6408490242266223,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.5456,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12257601320743561,
      "step": 310,
      "valid_targets_mean": 696.9,
      "valid_targets_min": 300
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.7541041516650534,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.5427,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.15626591444015503,
      "step": 315,
      "valid_targets_mean": 719.5,
      "valid_targets_min": 396
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.665090463756649,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.5352,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13162535429000854,
      "step": 320,
      "valid_targets_mean": 670.2,
      "valid_targets_min": 440
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.6366823920807368,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.5321,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13383950293064117,
      "step": 325,
      "valid_targets_mean": 810.5,
      "valid_targets_min": 394
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.6213548819315227,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.5265,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10977214574813843,
      "step": 330,
      "valid_targets_mean": 728.4,
      "valid_targets_min": 429
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.6547366858654076,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.5254,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1424219012260437,
      "step": 335,
      "valid_targets_mean": 800.0,
      "valid_targets_min": 452
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.6619785188617626,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.5306,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12859517335891724,
      "step": 340,
      "valid_targets_mean": 695.9,
      "valid_targets_min": 435
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.687101029751781,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.5184,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12573498487472534,
      "step": 345,
      "valid_targets_mean": 777.2,
      "valid_targets_min": 532
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.645409223244431,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.5215,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.118968665599823,
      "step": 350,
      "valid_targets_mean": 712.1,
      "valid_targets_min": 426
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.671387281197927,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.5203,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12927383184432983,
      "step": 355,
      "valid_targets_mean": 742.2,
      "valid_targets_min": 464
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.7041426331177664,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.5165,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13757221400737762,
      "step": 360,
      "valid_targets_mean": 762.4,
      "valid_targets_min": 397
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.5959539114433047,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.5197,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1234401986002922,
      "step": 365,
      "valid_targets_mean": 834.3,
      "valid_targets_min": 459
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.6531185712082775,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.5098,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1260676383972168,
      "step": 370,
      "valid_targets_mean": 682.6,
      "valid_targets_min": 395
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.6634208938956202,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.533,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1401265263557434,
      "step": 375,
      "valid_targets_mean": 758.4,
      "valid_targets_min": 325
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.6727520164532284,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.5271,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13311098515987396,
      "step": 380,
      "valid_targets_mean": 742.8,
      "valid_targets_min": 454
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.6427726949014289,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.5202,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13644330203533173,
      "step": 385,
      "valid_targets_mean": 789.9,
      "valid_targets_min": 428
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.6840033417316511,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.5308,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12673068046569824,
      "step": 390,
      "valid_targets_mean": 702.9,
      "valid_targets_min": 420
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.6563061162401826,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.5297,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1355002373456955,
      "step": 395,
      "valid_targets_mean": 729.9,
      "valid_targets_min": 526
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.641354939892573,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.5177,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12141138315200806,
      "step": 400,
      "valid_targets_mean": 733.4,
      "valid_targets_min": 316
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.8906863552430406,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.5165,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12363258004188538,
      "step": 405,
      "valid_targets_mean": 859.4,
      "valid_targets_min": 384
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.676141962708069,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.5358,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14006322622299194,
      "step": 410,
      "valid_targets_mean": 766.3,
      "valid_targets_min": 300
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.615596066669323,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.5215,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12042121589183807,
      "step": 415,
      "valid_targets_mean": 843.1,
      "valid_targets_min": 342
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.615943796645789,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.5144,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1382143646478653,
      "step": 420,
      "valid_targets_mean": 864.3,
      "valid_targets_min": 464
    },
    {
      "epoch": 2.7104,
      "grad_norm": 1.1772547291108486,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.535,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1458306610584259,
      "step": 425,
      "valid_targets_mean": 898.8,
      "valid_targets_min": 554
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.6393166323484019,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.5235,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13645440340042114,
      "step": 430,
      "valid_targets_mean": 740.2,
      "valid_targets_min": 391
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.6786636785060219,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.5337,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14586326479911804,
      "step": 435,
      "valid_targets_mean": 734.4,
      "valid_targets_min": 313
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.6138448927187343,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.5094,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14957866072654724,
      "step": 440,
      "valid_targets_mean": 983.4,
      "valid_targets_min": 476
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.6115317247558727,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.5215,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13156554102897644,
      "step": 445,
      "valid_targets_mean": 756.7,
      "valid_targets_min": 509
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.6480026309160363,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.5164,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13125823438167572,
      "step": 450,
      "valid_targets_mean": 749.1,
      "valid_targets_min": 384
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.6221700405662001,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.5289,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1366758495569229,
      "step": 455,
      "valid_targets_mean": 716.8,
      "valid_targets_min": 372
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.6835135386375543,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.5423,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1429317593574524,
      "step": 460,
      "valid_targets_mean": 738.8,
      "valid_targets_min": 370
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.6476761584366846,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.5177,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12608763575553894,
      "step": 465,
      "valid_targets_mean": 771.1,
      "valid_targets_min": 577
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.6705239133879689,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.5304,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.156068354845047,
      "step": 470,
      "valid_targets_mean": 891.4,
      "valid_targets_min": 436
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.6254648310969626,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.4947,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12908436357975006,
      "step": 475,
      "valid_targets_mean": 856.1,
      "valid_targets_min": 468
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.6835751653097379,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.5142,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14210891723632812,
      "step": 480,
      "valid_targets_mean": 753.6,
      "valid_targets_min": 416
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.6619131317365188,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.5013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12965424358844757,
      "step": 485,
      "valid_targets_mean": 726.4,
      "valid_targets_min": 444
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.6591313448070079,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.5063,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1259090006351471,
      "step": 490,
      "valid_targets_mean": 749.0,
      "valid_targets_min": 385
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.6569116597796409,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.5012,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1285029649734497,
      "step": 495,
      "valid_targets_mean": 837.4,
      "valid_targets_min": 496
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.6316910437995757,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.5031,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11511541903018951,
      "step": 500,
      "valid_targets_mean": 672.5,
      "valid_targets_min": 430
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.6389505729458541,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.5044,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12613186240196228,
      "step": 505,
      "valid_targets_mean": 796.8,
      "valid_targets_min": 310
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.6629145801717702,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.495,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11542559415102005,
      "step": 510,
      "valid_targets_mean": 727.7,
      "valid_targets_min": 439
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.634372579989713,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.5014,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12187671661376953,
      "step": 515,
      "valid_targets_mean": 753.3,
      "valid_targets_min": 482
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.6384070283786387,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.5008,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1421659290790558,
      "step": 520,
      "valid_targets_mean": 845.1,
      "valid_targets_min": 573
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.6332388154728119,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.4928,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11383464932441711,
      "step": 525,
      "valid_targets_mean": 718.1,
      "valid_targets_min": 397
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.6093799616673348,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.4926,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12254788726568222,
      "step": 530,
      "valid_targets_mean": 798.4,
      "valid_targets_min": 413
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.6455046460858423,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.5112,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13581742346286774,
      "step": 535,
      "valid_targets_mean": 846.3,
      "valid_targets_min": 520
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.631427790063437,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.5119,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1291504204273224,
      "step": 540,
      "valid_targets_mean": 806.6,
      "valid_targets_min": 501
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.6322674040782007,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.4964,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11973114311695099,
      "step": 545,
      "valid_targets_mean": 736.1,
      "valid_targets_min": 400
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.6441453581221164,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.5132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13486211001873016,
      "step": 550,
      "valid_targets_mean": 804.2,
      "valid_targets_min": 478
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.6425950808621582,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.5077,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12831318378448486,
      "step": 555,
      "valid_targets_mean": 724.8,
      "valid_targets_min": 512
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.6124823375637398,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.5038,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12071146070957184,
      "step": 560,
      "valid_targets_mean": 741.8,
      "valid_targets_min": 316
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.6205786442665068,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.5037,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10591184347867966,
      "step": 565,
      "valid_targets_mean": 709.2,
      "valid_targets_min": 480
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.6636478895992294,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.4993,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12058466672897339,
      "step": 570,
      "valid_targets_mean": 727.6,
      "valid_targets_min": 417
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.5971242481772463,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.4952,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13874907791614532,
      "step": 575,
      "valid_targets_mean": 860.1,
      "valid_targets_min": 527
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.615075544398069,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.5037,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11050853878259659,
      "step": 580,
      "valid_targets_mean": 743.3,
      "valid_targets_min": 468
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.6470427339088323,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.5064,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13953588902950287,
      "step": 585,
      "valid_targets_mean": 822.9,
      "valid_targets_min": 493
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.597610350393088,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.4967,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11674021184444427,
      "step": 590,
      "valid_targets_mean": 738.9,
      "valid_targets_min": 415
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.6338991358688321,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.5006,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12019048631191254,
      "step": 595,
      "valid_targets_mean": 707.8,
      "valid_targets_min": 442
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.6202165939689909,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.4968,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12805257737636566,
      "step": 600,
      "valid_targets_mean": 873.7,
      "valid_targets_min": 529
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.6714289302846277,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.5012,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12346205115318298,
      "step": 605,
      "valid_targets_mean": 609.8,
      "valid_targets_min": 416
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.6242136384257845,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.501,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11968934535980225,
      "step": 610,
      "valid_targets_mean": 689.3,
      "valid_targets_min": 443
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.6463661477202667,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.5053,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1359320878982544,
      "step": 615,
      "valid_targets_mean": 786.9,
      "valid_targets_min": 559
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.6491908587833649,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.4973,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14212754368782043,
      "step": 620,
      "valid_targets_mean": 861.1,
      "valid_targets_min": 507
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.6281434470761624,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.4975,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11817573010921478,
      "step": 625,
      "valid_targets_mean": 755.6,
      "valid_targets_min": 445
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.6490532684460447,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.4819,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11554375290870667,
      "step": 630,
      "valid_targets_mean": 724.1,
      "valid_targets_min": 367
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.6327445790214326,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.4894,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12214243412017822,
      "step": 635,
      "valid_targets_mean": 676.2,
      "valid_targets_min": 310
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.6158265147921352,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.4868,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13750189542770386,
      "step": 640,
      "valid_targets_mean": 834.1,
      "valid_targets_min": 555
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.627070881496122,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.4929,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11655047535896301,
      "step": 645,
      "valid_targets_mean": 674.8,
      "valid_targets_min": 411
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.6221973742999799,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.489,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12250182032585144,
      "step": 650,
      "valid_targets_mean": 665.8,
      "valid_targets_min": 394
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.6626884060775692,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.4741,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12655746936798096,
      "step": 655,
      "valid_targets_mean": 705.9,
      "valid_targets_min": 548
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.6224489306449362,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.4841,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12335214763879776,
      "step": 660,
      "valid_targets_mean": 784.9,
      "valid_targets_min": 525
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.6520842934256829,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.4911,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12841469049453735,
      "step": 665,
      "valid_targets_mean": 765.9,
      "valid_targets_min": 481
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.6423816689951226,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.4819,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10488501191139221,
      "step": 670,
      "valid_targets_mean": 732.0,
      "valid_targets_min": 450
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.6181615169830669,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.481,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12980340421199799,
      "step": 675,
      "valid_targets_mean": 780.8,
      "valid_targets_min": 466
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.6322677157647695,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.4964,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12320327758789062,
      "step": 680,
      "valid_targets_mean": 868.0,
      "valid_targets_min": 429
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.6043179861396448,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.4919,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1195220947265625,
      "step": 685,
      "valid_targets_mean": 728.2,
      "valid_targets_min": 471
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.640478901066271,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.4888,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12453019618988037,
      "step": 690,
      "valid_targets_mean": 716.9,
      "valid_targets_min": 362
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.6112657891179784,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.4972,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12277944386005402,
      "step": 695,
      "valid_targets_mean": 826.2,
      "valid_targets_min": 418
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.680661377919841,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.4926,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11709259450435638,
      "step": 700,
      "valid_targets_mean": 670.3,
      "valid_targets_min": 446
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.6067946877348196,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.4844,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11835283041000366,
      "step": 705,
      "valid_targets_mean": 833.8,
      "valid_targets_min": 535
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.6214656228794159,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.4955,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1317848563194275,
      "step": 710,
      "valid_targets_mean": 780.4,
      "valid_targets_min": 540
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.6101425589821468,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.4892,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1342540979385376,
      "step": 715,
      "valid_targets_mean": 813.8,
      "valid_targets_min": 592
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.6431480087423993,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.4892,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12438573688268661,
      "step": 720,
      "valid_targets_mean": 771.0,
      "valid_targets_min": 505
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.6403336841979095,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.502,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14776507019996643,
      "step": 725,
      "valid_targets_mean": 895.8,
      "valid_targets_min": 484
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.6038822017761685,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.4909,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11538829654455185,
      "step": 730,
      "valid_targets_mean": 747.9,
      "valid_targets_min": 428
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.6324119113515146,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.4979,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13889926671981812,
      "step": 735,
      "valid_targets_mean": 792.2,
      "valid_targets_min": 385
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.6324936257203374,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.487,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1103043407201767,
      "step": 740,
      "valid_targets_mean": 703.1,
      "valid_targets_min": 435
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.629729044732728,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.4892,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12301481515169144,
      "step": 745,
      "valid_targets_mean": 739.1,
      "valid_targets_min": 491
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.6274760958790774,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.4877,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11202463507652283,
      "step": 750,
      "valid_targets_mean": 644.5,
      "valid_targets_min": 356
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.6073341940792345,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.4851,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12768879532814026,
      "step": 755,
      "valid_targets_mean": 871.2,
      "valid_targets_min": 439
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.6300616469590965,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.4962,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10540274530649185,
      "step": 760,
      "valid_targets_mean": 732.8,
      "valid_targets_min": 458
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.607465565387994,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.4911,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.132490336894989,
      "step": 765,
      "valid_targets_mean": 854.8,
      "valid_targets_min": 435
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.5917313437808399,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.4864,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10872993618249893,
      "step": 770,
      "valid_targets_mean": 783.2,
      "valid_targets_min": 507
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.7403546016233938,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.4979,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12833866477012634,
      "step": 775,
      "valid_targets_mean": 677.8,
      "valid_targets_min": 428
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.6407918253430359,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.498,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12702913582324982,
      "step": 780,
      "valid_targets_mean": 762.4,
      "valid_targets_min": 411
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.2926809123761072,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.4833,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4620312452316284,
      "step": 785,
      "valid_targets_mean": 702.6,
      "valid_targets_min": 459
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.4620312452316284,
      "step": 785,
      "total_flos": 2.3852276139845222e+17,
      "train_loss": 0.5620896111628052,
      "train_runtime": 6728.981,
      "train_samples_per_second": 7.428,
      "train_steps_per_second": 0.117,
      "valid_targets_mean": 702.6,
      "valid_targets_min": 459
    }
  ],
  "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": 2.3852276139845222e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}