{
  "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": 5.442185453929223,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.6905,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.19166269898414612,
      "step": 5,
      "valid_targets_mean": 3772.9,
      "valid_targets_min": 1172
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.8095328022653705,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.6771,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.20748385787010193,
      "step": 10,
      "valid_targets_mean": 5537.4,
      "valid_targets_min": 684
    },
    {
      "epoch": 0.096,
      "grad_norm": 2.1200115743319388,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.639,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.16293901205062866,
      "step": 15,
      "valid_targets_mean": 4067.4,
      "valid_targets_min": 422
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.9257072627481844,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.5846,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1487666368484497,
      "step": 20,
      "valid_targets_mean": 3707.1,
      "valid_targets_min": 905
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7988850073269412,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.5754,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.19959388673305511,
      "step": 25,
      "valid_targets_mean": 5725.4,
      "valid_targets_min": 665
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.5331769320628815,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.5604,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12151822447776794,
      "step": 30,
      "valid_targets_mean": 3568.0,
      "valid_targets_min": 727
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.45282811415327257,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.5278,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11509503424167633,
      "step": 35,
      "valid_targets_mean": 3148.8,
      "valid_targets_min": 1224
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.4736002691754016,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.5098,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12758716940879822,
      "step": 40,
      "valid_targets_mean": 3358.9,
      "valid_targets_min": 811
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.31632295133430655,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.4781,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08016407489776611,
      "step": 45,
      "valid_targets_mean": 3059.4,
      "valid_targets_min": 790
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.348782345524125,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.4745,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13580970466136932,
      "step": 50,
      "valid_targets_mean": 4653.7,
      "valid_targets_min": 1002
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.25342842121431647,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.4667,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12497734278440475,
      "step": 55,
      "valid_targets_mean": 5929.0,
      "valid_targets_min": 737
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.2869586553254826,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.4442,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1034400537610054,
      "step": 60,
      "valid_targets_mean": 2641.9,
      "valid_targets_min": 651
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.3279100223436585,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.4432,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07635253667831421,
      "step": 65,
      "valid_targets_mean": 2183.4,
      "valid_targets_min": 624
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.27304560909035697,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.4394,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1170448511838913,
      "step": 70,
      "valid_targets_mean": 4851.8,
      "valid_targets_min": 920
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2666107311165606,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.4498,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12639470398426056,
      "step": 75,
      "valid_targets_mean": 5269.2,
      "valid_targets_min": 699
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.2772064811496187,
      "learning_rate": 4e-05,
      "loss": 0.4298,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0887710303068161,
      "step": 80,
      "valid_targets_mean": 2959.7,
      "valid_targets_min": 586
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.2544028411955918,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.4136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12488332390785217,
      "step": 85,
      "valid_targets_mean": 5450.7,
      "valid_targets_min": 693
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.27618561723531765,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.4111,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07022757828235626,
      "step": 90,
      "valid_targets_mean": 2691.6,
      "valid_targets_min": 878
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.29279387688561254,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.4225,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1592681109905243,
      "step": 95,
      "valid_targets_mean": 5913.5,
      "valid_targets_min": 789
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3076526359692288,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.4071,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08246097713708878,
      "step": 100,
      "valid_targets_mean": 3579.7,
      "valid_targets_min": 598
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.3196948304195173,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.4145,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07396609336137772,
      "step": 105,
      "valid_targets_mean": 2704.1,
      "valid_targets_min": 784
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.27897727552461854,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.4194,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1036178469657898,
      "step": 110,
      "valid_targets_mean": 4052.7,
      "valid_targets_min": 692
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.2999073837951396,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.4139,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08229044079780579,
      "step": 115,
      "valid_targets_mean": 3082.8,
      "valid_targets_min": 623
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.3358098936207653,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.4227,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14959537982940674,
      "step": 120,
      "valid_targets_mean": 4226.0,
      "valid_targets_min": 701
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2727171262170047,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.4236,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0725155621767044,
      "step": 125,
      "valid_targets_mean": 2769.6,
      "valid_targets_min": 632
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.2728697189925512,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.4219,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09483879804611206,
      "step": 130,
      "valid_targets_mean": 3119.8,
      "valid_targets_min": 1146
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.27493747655842643,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.4051,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07629527151584625,
      "step": 135,
      "valid_targets_mean": 4574.4,
      "valid_targets_min": 574
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.273673015320205,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.4214,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09848488122224808,
      "step": 140,
      "valid_targets_mean": 4396.8,
      "valid_targets_min": 779
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.2835354881684314,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.4352,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06863738596439362,
      "step": 145,
      "valid_targets_mean": 2600.9,
      "valid_targets_min": 884
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.24549306111748626,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.415,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10824453830718994,
      "step": 150,
      "valid_targets_mean": 5110.8,
      "valid_targets_min": 725
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.26293935756408227,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.4054,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08755393326282501,
      "step": 155,
      "valid_targets_mean": 4606.3,
      "valid_targets_min": 771
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.28805917974478973,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.3989,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08316570520401001,
      "step": 160,
      "valid_targets_mean": 3181.9,
      "valid_targets_min": 825
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.26820639034503513,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.3909,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08125826716423035,
      "step": 165,
      "valid_targets_mean": 4078.8,
      "valid_targets_min": 956
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.2799230229956331,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.3972,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09274695068597794,
      "step": 170,
      "valid_targets_mean": 3781.8,
      "valid_targets_min": 983
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.26205953485259365,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.4088,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07772032916545868,
      "step": 175,
      "valid_targets_mean": 2916.4,
      "valid_targets_min": 571
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.2637599374407512,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.3943,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09445404261350632,
      "step": 180,
      "valid_targets_mean": 4496.7,
      "valid_targets_min": 800
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.2932291816861461,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.3781,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09436005353927612,
      "step": 185,
      "valid_targets_mean": 3035.1,
      "valid_targets_min": 693
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.27550995248750926,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.3979,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10137677937746048,
      "step": 190,
      "valid_targets_mean": 4430.1,
      "valid_targets_min": 368
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.2639313244357049,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.3969,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08653558790683746,
      "step": 195,
      "valid_targets_mean": 4251.3,
      "valid_targets_min": 563
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.3362952524470988,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.3867,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11065793037414551,
      "step": 200,
      "valid_targets_mean": 4629.8,
      "valid_targets_min": 706
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.2650542140015465,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.3858,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10903624445199966,
      "step": 205,
      "valid_targets_mean": 4045.3,
      "valid_targets_min": 841
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.3676272929156744,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.3987,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0916772335767746,
      "step": 210,
      "valid_targets_mean": 2358.5,
      "valid_targets_min": 489
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.29610462749256095,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.3786,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14761430025100708,
      "step": 215,
      "valid_targets_mean": 5930.5,
      "valid_targets_min": 512
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.2949118655484367,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.3777,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10777334123849869,
      "step": 220,
      "valid_targets_mean": 4821.2,
      "valid_targets_min": 632
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.2600378365295041,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.3941,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.078704833984375,
      "step": 225,
      "valid_targets_mean": 4813.6,
      "valid_targets_min": 695
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.24938523801103532,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.3786,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08179652690887451,
      "step": 230,
      "valid_targets_mean": 4133.4,
      "valid_targets_min": 662
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.29863052946621094,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.3749,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10285639762878418,
      "step": 235,
      "valid_targets_mean": 3773.5,
      "valid_targets_min": 661
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.26854744897753935,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.3791,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09314319491386414,
      "step": 240,
      "valid_targets_mean": 3503.4,
      "valid_targets_min": 690
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.354465503459906,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.3807,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07929793000221252,
      "step": 245,
      "valid_targets_mean": 2074.6,
      "valid_targets_min": 808
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.2527042035107904,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.4069,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10088837146759033,
      "step": 250,
      "valid_targets_mean": 5246.9,
      "valid_targets_min": 1264
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.2589368745471832,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.3851,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10473264008760452,
      "step": 255,
      "valid_targets_mean": 5016.7,
      "valid_targets_min": 1757
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.28003847497063306,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.3998,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09361633658409119,
      "step": 260,
      "valid_targets_mean": 3909.4,
      "valid_targets_min": 785
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.2943787611971499,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.3774,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08121328055858612,
      "step": 265,
      "valid_targets_mean": 3470.5,
      "valid_targets_min": 780
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.264050322073341,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.3797,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11807336658239365,
      "step": 270,
      "valid_targets_mean": 4398.0,
      "valid_targets_min": 735
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.2580218568904529,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.3618,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.096112921833992,
      "step": 275,
      "valid_targets_mean": 5801.4,
      "valid_targets_min": 547
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.30673536056998396,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.3881,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08830038458108902,
      "step": 280,
      "valid_targets_mean": 3337.2,
      "valid_targets_min": 1124
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.2835089845775815,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.4036,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1227225810289383,
      "step": 285,
      "valid_targets_mean": 4012.8,
      "valid_targets_min": 861
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.2523256388128284,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.3794,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08101785182952881,
      "step": 290,
      "valid_targets_mean": 3038.2,
      "valid_targets_min": 823
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.26924124501345237,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.3806,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10177544504404068,
      "step": 295,
      "valid_targets_mean": 5270.2,
      "valid_targets_min": 820
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.28130372404313303,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.3778,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07558295130729675,
      "step": 300,
      "valid_targets_mean": 4471.8,
      "valid_targets_min": 600
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.28205364927715976,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.387,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12669092416763306,
      "step": 305,
      "valid_targets_mean": 5147.2,
      "valid_targets_min": 578
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.272400846069077,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.3724,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11067002266645432,
      "step": 310,
      "valid_targets_mean": 3779.5,
      "valid_targets_min": 955
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.26765963746969057,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.3865,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.100395068526268,
      "step": 315,
      "valid_targets_mean": 4568.2,
      "valid_targets_min": 792
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.27706186342808137,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.3781,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.13206183910369873,
      "step": 320,
      "valid_targets_mean": 4657.1,
      "valid_targets_min": 834
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.2661407560456541,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.3491,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09652642160654068,
      "step": 325,
      "valid_targets_mean": 5487.9,
      "valid_targets_min": 1218
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.31701150346198914,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.37,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08779852092266083,
      "step": 330,
      "valid_targets_mean": 3176.9,
      "valid_targets_min": 762
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.27075908827848955,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.369,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09712551534175873,
      "step": 335,
      "valid_targets_mean": 4385.8,
      "valid_targets_min": 867
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.2492639945727957,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.381,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08740992844104767,
      "step": 340,
      "valid_targets_mean": 4696.7,
      "valid_targets_min": 796
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.30232927389846304,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.3761,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11960571259260178,
      "step": 345,
      "valid_targets_mean": 4552.9,
      "valid_targets_min": 610
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.28463853698825686,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.3553,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0824894830584526,
      "step": 350,
      "valid_targets_mean": 3428.7,
      "valid_targets_min": 645
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.2619547638651442,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.3674,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12523266673088074,
      "step": 355,
      "valid_targets_mean": 5121.7,
      "valid_targets_min": 827
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.2826176494689443,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.3698,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10490305721759796,
      "step": 360,
      "valid_targets_mean": 5392.1,
      "valid_targets_min": 1163
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.26281886388054504,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.3698,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08223024010658264,
      "step": 365,
      "valid_targets_mean": 4496.4,
      "valid_targets_min": 891
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.28520427596756726,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.3659,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07992172241210938,
      "step": 370,
      "valid_targets_mean": 3386.4,
      "valid_targets_min": 649
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.2925895212634724,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.3579,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09350573271512985,
      "step": 375,
      "valid_targets_mean": 4317.8,
      "valid_targets_min": 546
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.25993826727846614,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.3574,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10308262705802917,
      "step": 380,
      "valid_targets_mean": 5797.9,
      "valid_targets_min": 1315
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.26929118383315276,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.3619,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11039996147155762,
      "step": 385,
      "valid_targets_mean": 4797.1,
      "valid_targets_min": 923
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.2905654079690829,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.3756,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0772155225276947,
      "step": 390,
      "valid_targets_mean": 2625.3,
      "valid_targets_min": 715
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.2586265822283464,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.3511,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08459736406803131,
      "step": 395,
      "valid_targets_mean": 5125.2,
      "valid_targets_min": 665
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.24881513909392297,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.3666,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1012953445315361,
      "step": 400,
      "valid_targets_mean": 5529.9,
      "valid_targets_min": 575
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.2477813914251035,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.3786,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10448640584945679,
      "step": 405,
      "valid_targets_mean": 5946.4,
      "valid_targets_min": 1290
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.2861168060044399,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.3707,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09149201959371567,
      "step": 410,
      "valid_targets_mean": 4533.5,
      "valid_targets_min": 707
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.2714708833558892,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.3788,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0974833071231842,
      "step": 415,
      "valid_targets_mean": 4389.8,
      "valid_targets_min": 422
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.3115246578077533,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.3806,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10238995403051376,
      "step": 420,
      "valid_targets_mean": 3236.0,
      "valid_targets_min": 799
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.30112620981655347,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.3785,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09492282569408417,
      "step": 425,
      "valid_targets_mean": 3743.2,
      "valid_targets_min": 766
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.2812525502947618,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.3818,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.1166185587644577,
      "step": 430,
      "valid_targets_mean": 4319.6,
      "valid_targets_min": 1012
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.24849187677677648,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.3497,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08945291489362717,
      "step": 435,
      "valid_targets_mean": 5814.7,
      "valid_targets_min": 784
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.2631331467696701,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.3797,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10857097804546356,
      "step": 440,
      "valid_targets_mean": 4672.4,
      "valid_targets_min": 1115
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.26773590398116837,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.3578,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06477227061986923,
      "step": 445,
      "valid_targets_mean": 3280.0,
      "valid_targets_min": 463
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.2593193114272387,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.3645,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08953142166137695,
      "step": 450,
      "valid_targets_mean": 5166.1,
      "valid_targets_min": 610
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.2617695948022818,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.368,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08089570701122284,
      "step": 455,
      "valid_targets_mean": 3814.2,
      "valid_targets_min": 706
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.2357322241148187,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.346,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08805036544799805,
      "step": 460,
      "valid_targets_mean": 5983.0,
      "valid_targets_min": 733
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.26747148060678083,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.3777,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11624380946159363,
      "step": 465,
      "valid_targets_mean": 5326.9,
      "valid_targets_min": 807
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.2676459351712619,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.3754,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10637401044368744,
      "step": 470,
      "valid_targets_mean": 4809.9,
      "valid_targets_min": 683
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.2527197970631467,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.3763,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08800717443227768,
      "step": 475,
      "valid_targets_mean": 4288.8,
      "valid_targets_min": 523
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.26943933773942935,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.3633,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08747033029794693,
      "step": 480,
      "valid_targets_mean": 3784.7,
      "valid_targets_min": 707
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.2929312520277543,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.3534,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.062183164060115814,
      "step": 485,
      "valid_targets_mean": 1691.6,
      "valid_targets_min": 563
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.2637865324511221,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.3575,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09510040283203125,
      "step": 490,
      "valid_targets_mean": 3853.3,
      "valid_targets_min": 709
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.30734784710913193,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.363,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09190519154071808,
      "step": 495,
      "valid_targets_mean": 3746.6,
      "valid_targets_min": 833
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.24787331839944854,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.347,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08131758123636246,
      "step": 500,
      "valid_targets_mean": 4313.7,
      "valid_targets_min": 990
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.24383293934449407,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.3593,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06853271275758743,
      "step": 505,
      "valid_targets_mean": 4191.8,
      "valid_targets_min": 659
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.2965221142379322,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.3468,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07949408888816833,
      "step": 510,
      "valid_targets_mean": 3390.6,
      "valid_targets_min": 762
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.2492892033414412,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.3505,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07512718439102173,
      "step": 515,
      "valid_targets_mean": 4176.8,
      "valid_targets_min": 830
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.267290487021705,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.3525,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06713395565748215,
      "step": 520,
      "valid_targets_mean": 3405.8,
      "valid_targets_min": 732
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.2551948302077571,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.3476,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07601924240589142,
      "step": 525,
      "valid_targets_mean": 3619.9,
      "valid_targets_min": 905
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.25252309788307625,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.3549,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07913119345903397,
      "step": 530,
      "valid_targets_mean": 3200.8,
      "valid_targets_min": 661
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.3155156588956853,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.3656,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09287755191326141,
      "step": 535,
      "valid_targets_mean": 3214.9,
      "valid_targets_min": 747
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.2766133077094186,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.3626,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09605400264263153,
      "step": 540,
      "valid_targets_mean": 4218.4,
      "valid_targets_min": 1095
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.27258472577367454,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.3696,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08149433881044388,
      "step": 545,
      "valid_targets_mean": 4192.5,
      "valid_targets_min": 747
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.25426394436556843,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.3589,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08685190975666046,
      "step": 550,
      "valid_targets_mean": 4254.9,
      "valid_targets_min": 757
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.27153496157844426,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.3657,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10605305433273315,
      "step": 555,
      "valid_targets_mean": 4210.9,
      "valid_targets_min": 1488
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.3537526334104841,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.3611,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09752020239830017,
      "step": 560,
      "valid_targets_mean": 2972.8,
      "valid_targets_min": 794
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.2651341936936712,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.3505,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07173003256320953,
      "step": 565,
      "valid_targets_mean": 3701.8,
      "valid_targets_min": 734
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.25545676170795567,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.3618,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07571163773536682,
      "step": 570,
      "valid_targets_mean": 4587.5,
      "valid_targets_min": 897
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.25947716758369005,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.3568,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10017608106136322,
      "step": 575,
      "valid_targets_mean": 4887.9,
      "valid_targets_min": 920
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.24666029210741366,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.3604,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07228102535009384,
      "step": 580,
      "valid_targets_mean": 3624.2,
      "valid_targets_min": 682
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.30422926479905615,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.3529,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0982704907655716,
      "step": 585,
      "valid_targets_mean": 4152.2,
      "valid_targets_min": 872
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.2848406589807159,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.3565,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06402648240327835,
      "step": 590,
      "valid_targets_mean": 2856.8,
      "valid_targets_min": 1025
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.3175945368962988,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.3625,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07294386625289917,
      "step": 595,
      "valid_targets_mean": 2645.3,
      "valid_targets_min": 1036
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.2662096619930337,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.3669,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07283228635787964,
      "step": 600,
      "valid_targets_mean": 3706.9,
      "valid_targets_min": 1047
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.2594850403272501,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.3504,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10935956239700317,
      "step": 605,
      "valid_targets_mean": 5242.9,
      "valid_targets_min": 665
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.24713702492526102,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.3458,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06710461527109146,
      "step": 610,
      "valid_targets_mean": 4347.1,
      "valid_targets_min": 1322
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.2897155890131026,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.3625,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08740752935409546,
      "step": 615,
      "valid_targets_mean": 3607.1,
      "valid_targets_min": 600
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.24379821604551838,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.3501,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0911904126405716,
      "step": 620,
      "valid_targets_mean": 4610.9,
      "valid_targets_min": 859
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.3154382094028816,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.3646,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.105601966381073,
      "step": 625,
      "valid_targets_mean": 4914.9,
      "valid_targets_min": 537
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.2728978101785218,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.3477,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06833998113870621,
      "step": 630,
      "valid_targets_mean": 4309.9,
      "valid_targets_min": 612
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.2667331818881737,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.3528,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08249158412218094,
      "step": 635,
      "valid_targets_mean": 3725.8,
      "valid_targets_min": 573
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.2653027077039279,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.3516,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11303485184907913,
      "step": 640,
      "valid_targets_mean": 5256.0,
      "valid_targets_min": 1442
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.26248507735732046,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.338,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06639324128627777,
      "step": 645,
      "valid_targets_mean": 3318.1,
      "valid_targets_min": 679
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.27929839995512357,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.3415,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05943456292152405,
      "step": 650,
      "valid_targets_mean": 2786.4,
      "valid_targets_min": 777
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.2586565562817669,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.3615,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08127006888389587,
      "step": 655,
      "valid_targets_mean": 3574.6,
      "valid_targets_min": 840
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.26935723611482854,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.3525,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07011975347995758,
      "step": 660,
      "valid_targets_mean": 2604.6,
      "valid_targets_min": 726
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.2637733286008412,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.3499,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09666310250759125,
      "step": 665,
      "valid_targets_mean": 5042.1,
      "valid_targets_min": 544
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.32987725919571403,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.3476,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08155728876590729,
      "step": 670,
      "valid_targets_mean": 3523.2,
      "valid_targets_min": 953
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.2715487393904151,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.3637,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.09598340094089508,
      "step": 675,
      "valid_targets_mean": 4457.3,
      "valid_targets_min": 692
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.2812120700507707,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.3644,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11278092116117477,
      "step": 680,
      "valid_targets_mean": 4421.0,
      "valid_targets_min": 730
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.2579264284472369,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.35,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07923750579357147,
      "step": 685,
      "valid_targets_mean": 3840.6,
      "valid_targets_min": 759
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.2534367716071427,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.3572,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.06656123697757721,
      "step": 690,
      "valid_targets_mean": 2848.0,
      "valid_targets_min": 892
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.2614144350234106,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.3663,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08406713604927063,
      "step": 695,
      "valid_targets_mean": 4129.6,
      "valid_targets_min": 787
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.26251126298421695,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.3609,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07219307124614716,
      "step": 700,
      "valid_targets_mean": 3461.7,
      "valid_targets_min": 1063
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.2818927782701796,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.3505,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.14388665556907654,
      "step": 705,
      "valid_targets_mean": 5650.9,
      "valid_targets_min": 662
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.25013055924766153,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.3463,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.078327476978302,
      "step": 710,
      "valid_targets_mean": 3948.1,
      "valid_targets_min": 1196
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.23856915999519057,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.356,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08584430068731308,
      "step": 715,
      "valid_targets_mean": 4553.8,
      "valid_targets_min": 659
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.29180461161366616,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.3718,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10328815877437592,
      "step": 720,
      "valid_targets_mean": 4305.4,
      "valid_targets_min": 560
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.25445248125988607,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.3412,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08515608310699463,
      "step": 725,
      "valid_targets_mean": 3890.6,
      "valid_targets_min": 784
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.2689653858867643,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.3677,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.10580168664455414,
      "step": 730,
      "valid_targets_mean": 3736.6,
      "valid_targets_min": 1083
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.28761622388224484,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.3566,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.12114425003528595,
      "step": 735,
      "valid_targets_mean": 4689.3,
      "valid_targets_min": 940
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.24365172717782563,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.3328,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0749347135424614,
      "step": 740,
      "valid_targets_mean": 4185.6,
      "valid_targets_min": 653
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.2283492870630903,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.3434,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07939842343330383,
      "step": 745,
      "valid_targets_mean": 5340.9,
      "valid_targets_min": 1265
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.29806755489113096,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.3582,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.11634281277656555,
      "step": 750,
      "valid_targets_mean": 4905.9,
      "valid_targets_min": 695
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.2671590330780972,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.3383,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.05478291213512421,
      "step": 755,
      "valid_targets_mean": 2623.4,
      "valid_targets_min": 595
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.2706109149704598,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.3525,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07864299416542053,
      "step": 760,
      "valid_targets_mean": 3891.2,
      "valid_targets_min": 668
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.2759378816487873,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.3635,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08948777616024017,
      "step": 765,
      "valid_targets_mean": 3952.4,
      "valid_targets_min": 644
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.24354389118800948,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.3437,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.07612688839435577,
      "step": 770,
      "valid_targets_mean": 4240.5,
      "valid_targets_min": 1037
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.2613977673526795,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.3511,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0692347064614296,
      "step": 775,
      "valid_targets_mean": 3165.6,
      "valid_targets_min": 703
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.25143698509914547,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.3495,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.08967740833759308,
      "step": 780,
      "valid_targets_mean": 4549.7,
      "valid_targets_min": 551
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5494204534303122,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.3527,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.36648064851760864,
      "step": 785,
      "valid_targets_mean": 3627.8,
      "valid_targets_min": 954
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.36648064851760864,
      "step": 785,
      "total_flos": 1.0910552897523548e+18,
      "train_loss": 0.3873556343613157,
      "train_runtime": 17001.6953,
      "train_samples_per_second": 2.941,
      "train_steps_per_second": 0.046,
      "valid_targets_mean": 3627.8,
      "valid_targets_min": 954
    }
  ],
  "logging_steps": 5,
  "max_steps": 785,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0910552897523548e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}