{
  "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": 3.918294621561185,
      "learning_rate": 2.0253164556962026e-06,
      "loss": 0.1164,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.022308267652988434,
      "step": 5,
      "valid_targets_mean": 1546.6,
      "valid_targets_min": 359
    },
    {
      "epoch": 0.064,
      "grad_norm": 3.1022539368816218,
      "learning_rate": 4.556962025316456e-06,
      "loss": 0.096,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.021156825125217438,
      "step": 10,
      "valid_targets_mean": 2994.7,
      "valid_targets_min": 322
    },
    {
      "epoch": 0.096,
      "grad_norm": 1.4715649539860216,
      "learning_rate": 7.08860759493671e-06,
      "loss": 0.0641,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.012252530083060265,
      "step": 15,
      "valid_targets_mean": 1052.1,
      "valid_targets_min": 358
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.9245936851089113,
      "learning_rate": 9.620253164556963e-06,
      "loss": 0.0365,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.007999857887625694,
      "step": 20,
      "valid_targets_mean": 2156.5,
      "valid_targets_min": 504
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.41441879310754387,
      "learning_rate": 1.2151898734177216e-05,
      "loss": 0.0232,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.006338654085993767,
      "step": 25,
      "valid_targets_mean": 2148.5,
      "valid_targets_min": 409
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.07787502090023249,
      "learning_rate": 1.468354430379747e-05,
      "loss": 0.0164,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0031789226923137903,
      "step": 30,
      "valid_targets_mean": 2205.9,
      "valid_targets_min": 485
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.5418680698018585,
      "learning_rate": 1.7215189873417723e-05,
      "loss": 0.0151,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003683611052110791,
      "step": 35,
      "valid_targets_mean": 2321.8,
      "valid_targets_min": 383
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.4444024344678225,
      "learning_rate": 1.974683544303798e-05,
      "loss": 0.0144,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003324975026771426,
      "step": 40,
      "valid_targets_mean": 1938.1,
      "valid_targets_min": 366
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.10871896770220735,
      "learning_rate": 2.2278481012658228e-05,
      "loss": 0.014,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0039204806089401245,
      "step": 45,
      "valid_targets_mean": 2420.4,
      "valid_targets_min": 473
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.32014810347174005,
      "learning_rate": 2.481012658227848e-05,
      "loss": 0.0142,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004062778316438198,
      "step": 50,
      "valid_targets_mean": 2242.8,
      "valid_targets_min": 272
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.30141213859659566,
      "learning_rate": 2.7341772151898737e-05,
      "loss": 0.0145,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003315784502774477,
      "step": 55,
      "valid_targets_mean": 1512.0,
      "valid_targets_min": 367
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.15386126304127462,
      "learning_rate": 2.987341772151899e-05,
      "loss": 0.014,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0029263454489409924,
      "step": 60,
      "valid_targets_mean": 1733.6,
      "valid_targets_min": 286
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.5055952269665289,
      "learning_rate": 3.240506329113924e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0036597000434994698,
      "step": 65,
      "valid_targets_mean": 2809.1,
      "valid_targets_min": 468
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.32446432643287154,
      "learning_rate": 3.49367088607595e-05,
      "loss": 0.014,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035838079638779163,
      "step": 70,
      "valid_targets_mean": 1804.4,
      "valid_targets_min": 308
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.18698877660157018,
      "learning_rate": 3.746835443037975e-05,
      "loss": 0.0141,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0034235180355608463,
      "step": 75,
      "valid_targets_mean": 1839.8,
      "valid_targets_min": 459
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.20681790983461557,
      "learning_rate": 4e-05,
      "loss": 0.0139,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003846520557999611,
      "step": 80,
      "valid_targets_mean": 2235.2,
      "valid_targets_min": 314
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.15437937807973956,
      "learning_rate": 3.999504991751045e-05,
      "loss": 0.0139,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0029371827840805054,
      "step": 85,
      "valid_targets_mean": 1222.9,
      "valid_targets_min": 242
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.1973268940520538,
      "learning_rate": 3.9980202120373464e-05,
      "loss": 0.0141,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003376382403075695,
      "step": 90,
      "valid_targets_mean": 2208.1,
      "valid_targets_min": 364
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.465038543804949,
      "learning_rate": 3.995546395837111e-05,
      "loss": 0.0145,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0034364010207355022,
      "step": 95,
      "valid_targets_mean": 2251.0,
      "valid_targets_min": 383
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.26792572631707545,
      "learning_rate": 3.992084767709763e-05,
      "loss": 0.0133,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003912950400263071,
      "step": 100,
      "valid_targets_mean": 3987.7,
      "valid_targets_min": 400
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.15703686602535893,
      "learning_rate": 3.987637041189781e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0034670978784561157,
      "step": 105,
      "valid_targets_mean": 2150.7,
      "valid_targets_min": 385
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.21984043572620518,
      "learning_rate": 3.982205417938482e-05,
      "loss": 0.0135,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0026204963214695454,
      "step": 110,
      "valid_targets_mean": 2346.5,
      "valid_targets_min": 368
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.15758946282978825,
      "learning_rate": 3.975792586654179e-05,
      "loss": 0.0139,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0031727757304906845,
      "step": 115,
      "valid_targets_mean": 1853.3,
      "valid_targets_min": 390
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.2940726341145156,
      "learning_rate": 3.968401721741259e-05,
      "loss": 0.0138,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004647557158023119,
      "step": 120,
      "valid_targets_mean": 2631.7,
      "valid_targets_min": 531
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.35025445696545787,
      "learning_rate": 3.960036481738819e-05,
      "loss": 0.0133,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0036999774165451527,
      "step": 125,
      "valid_targets_mean": 2538.4,
      "valid_targets_min": 490
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.2912902217578897,
      "learning_rate": 3.950701007509667e-05,
      "loss": 0.0134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0030563287436962128,
      "step": 130,
      "valid_targets_mean": 2333.6,
      "valid_targets_min": 374
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.07689521121143372,
      "learning_rate": 3.940399920190552e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0034338519908487797,
      "step": 135,
      "valid_targets_mean": 1838.0,
      "valid_targets_min": 417
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.18505137728409618,
      "learning_rate": 3.92913831890467e-05,
      "loss": 0.0134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004238358698785305,
      "step": 140,
      "valid_targets_mean": 2815.8,
      "valid_targets_min": 438
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.13029451826995153,
      "learning_rate": 3.916921778237556e-05,
      "loss": 0.0131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0031057405285537243,
      "step": 145,
      "valid_targets_mean": 1753.0,
      "valid_targets_min": 434
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.05565854599085797,
      "learning_rate": 3.903756345477612e-05,
      "loss": 0.0126,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002463353332132101,
      "step": 150,
      "valid_targets_mean": 1927.5,
      "valid_targets_min": 381
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.29141120781386237,
      "learning_rate": 3.889648537622657e-05,
      "loss": 0.0138,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002806419041007757,
      "step": 155,
      "valid_targets_mean": 1992.1,
      "valid_targets_min": 330
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.19535705221000835,
      "learning_rate": 3.874605338153952e-05,
      "loss": 0.0135,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0030665425583720207,
      "step": 160,
      "valid_targets_mean": 2763.3,
      "valid_targets_min": 411
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.10203530200801582,
      "learning_rate": 3.8586341935793265e-05,
      "loss": 0.0134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003486543893814087,
      "step": 165,
      "valid_targets_mean": 2278.7,
      "valid_targets_min": 379
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.12637754319053376,
      "learning_rate": 3.841743009747089e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003909777849912643,
      "step": 170,
      "valid_targets_mean": 3117.1,
      "valid_targets_min": 461
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.09908220885707889,
      "learning_rate": 3.8239401479325714e-05,
      "loss": 0.0135,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003633004380390048,
      "step": 175,
      "valid_targets_mean": 2515.3,
      "valid_targets_min": 388
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.10395096062217227,
      "learning_rate": 3.8052344206992276e-05,
      "loss": 0.0141,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003890944877639413,
      "step": 180,
      "valid_targets_mean": 1996.1,
      "valid_targets_min": 323
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.19895865258829665,
      "learning_rate": 3.7856350875363396e-05,
      "loss": 0.0139,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004547910299152136,
      "step": 185,
      "valid_targets_mean": 3006.2,
      "valid_targets_min": 347
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.11906444217251856,
      "learning_rate": 3.765151850275497e-05,
      "loss": 0.0134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035077473148703575,
      "step": 190,
      "valid_targets_mean": 2148.6,
      "valid_targets_min": 365
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.11223996781207712,
      "learning_rate": 3.7437948482881104e-05,
      "loss": 0.0133,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0027180458419024944,
      "step": 195,
      "valid_targets_mean": 1303.2,
      "valid_targets_min": 362
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.10961859454658451,
      "learning_rate": 3.721574653466336e-05,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002405869774520397,
      "step": 200,
      "valid_targets_mean": 2209.1,
      "valid_targets_min": 318
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.09725108059356587,
      "learning_rate": 3.698502264989903e-05,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003792133182287216,
      "step": 205,
      "valid_targets_mean": 2783.3,
      "valid_targets_min": 404
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.1305426996775664,
      "learning_rate": 3.674589103881432e-05,
      "loss": 0.0137,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0029035669285804033,
      "step": 210,
      "valid_targets_mean": 2007.8,
      "valid_targets_min": 525
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.21164045948055357,
      "learning_rate": 3.64984700735293e-05,
      "loss": 0.0137,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035701056476682425,
      "step": 215,
      "valid_targets_mean": 2523.5,
      "valid_targets_min": 363
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.07778881815645956,
      "learning_rate": 3.624288222946273e-05,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002559133805334568,
      "step": 220,
      "valid_targets_mean": 2254.9,
      "valid_targets_min": 393
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.09463892897238466,
      "learning_rate": 3.597925402470578e-05,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0028019435703754425,
      "step": 225,
      "valid_targets_mean": 2016.5,
      "valid_targets_min": 340
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.12404721009153415,
      "learning_rate": 3.570771595739445e-05,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0034294212237000465,
      "step": 230,
      "valid_targets_mean": 3196.8,
      "valid_targets_min": 357
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.09043514696164454,
      "learning_rate": 3.5428402441111964e-05,
      "loss": 0.0125,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0027854335494339466,
      "step": 235,
      "valid_targets_mean": 1582.8,
      "valid_targets_min": 477
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.05949281217734789,
      "learning_rate": 3.5141451738352936e-05,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0025357198901474476,
      "step": 240,
      "valid_targets_mean": 1867.8,
      "valid_targets_min": 403
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.044714337962087035,
      "learning_rate": 3.4847005892082266e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0037834374234080315,
      "step": 245,
      "valid_targets_mean": 1690.1,
      "valid_targets_min": 341
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.038045652571088065,
      "learning_rate": 3.454521065542273e-05,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0028451373800635338,
      "step": 250,
      "valid_targets_mean": 1950.0,
      "valid_targets_min": 412
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.019818931343667405,
      "learning_rate": 3.423621541950597e-05,
      "loss": 0.013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003329153172671795,
      "step": 255,
      "valid_targets_mean": 2559.8,
      "valid_targets_min": 550
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.16300453293326536,
      "learning_rate": 3.3920173139522664e-05,
      "loss": 0.0133,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0028890548273921013,
      "step": 260,
      "valid_targets_mean": 1251.0,
      "valid_targets_min": 395
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.11512348406030073,
      "learning_rate": 3.35972402590084e-05,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003410006407648325,
      "step": 265,
      "valid_targets_mean": 2369.7,
      "valid_targets_min": 425
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.21291297819192026,
      "learning_rate": 3.326757663240291e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0037192627787590027,
      "step": 270,
      "valid_targets_mean": 2423.2,
      "valid_targets_min": 439
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.1535791523678035,
      "learning_rate": 3.293134544592073e-05,
      "loss": 0.0131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035847551189363003,
      "step": 275,
      "valid_targets_mean": 2537.2,
      "valid_targets_min": 410
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.06318409988742545,
      "learning_rate": 3.258871313677274e-05,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0032514675986021757,
      "step": 280,
      "valid_targets_mean": 2719.9,
      "valid_targets_min": 369
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.03803030364927751,
      "learning_rate": 3.2239849310778316e-05,
      "loss": 0.0133,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0033862553536891937,
      "step": 285,
      "valid_targets_mean": 2779.4,
      "valid_targets_min": 354
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.24622775027917343,
      "learning_rate": 3.188492665840909e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0033008321188390255,
      "step": 290,
      "valid_targets_mean": 1894.0,
      "valid_targets_min": 373
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.034851967876114075,
      "learning_rate": 3.1524120869305726e-05,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0024069349747151136,
      "step": 295,
      "valid_targets_mean": 1592.9,
      "valid_targets_min": 330
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.1217042513015231,
      "learning_rate": 3.11576105453101e-05,
      "loss": 0.013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035706735216081142,
      "step": 300,
      "valid_targets_mean": 3020.6,
      "valid_targets_min": 468
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.05625414147657382,
      "learning_rate": 3.0785577112055916e-05,
      "loss": 0.0137,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004759094212204218,
      "step": 305,
      "valid_targets_mean": 2956.2,
      "valid_targets_min": 474
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.027062493107324948,
      "learning_rate": 3.040820472916153e-05,
      "loss": 0.0134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003156646154820919,
      "step": 310,
      "valid_targets_mean": 2193.2,
      "valid_targets_min": 408
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.14282826653594197,
      "learning_rate": 3.002568019906939e-05,
      "loss": 0.0148,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035651333164423704,
      "step": 315,
      "valid_targets_mean": 1683.8,
      "valid_targets_min": 288
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.2321934196291223,
      "learning_rate": 2.963819287457733e-05,
      "loss": 0.0135,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0026549678295850754,
      "step": 320,
      "valid_targets_mean": 1795.3,
      "valid_targets_min": 278
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.29087392863489386,
      "learning_rate": 2.924593456510733e-05,
      "loss": 0.0138,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004393232986330986,
      "step": 325,
      "valid_targets_mean": 2332.2,
      "valid_targets_min": 495
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.677204943133956,
      "learning_rate": 2.8849099441758306e-05,
      "loss": 0.014,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003810121212154627,
      "step": 330,
      "valid_targets_mean": 1778.2,
      "valid_targets_min": 359
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.20194970256381728,
      "learning_rate": 2.844788394118979e-05,
      "loss": 0.0143,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0025422400794923306,
      "step": 335,
      "valid_targets_mean": 1369.6,
      "valid_targets_min": 349
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.4839028788836421,
      "learning_rate": 2.8042486668384164e-05,
      "loss": 0.0137,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003389551304280758,
      "step": 340,
      "valid_targets_mean": 2752.8,
      "valid_targets_min": 510
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.2749371493800895,
      "learning_rate": 2.7633108298335582e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004174442961812019,
      "step": 345,
      "valid_targets_mean": 2299.1,
      "valid_targets_min": 397
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.293211077809033,
      "learning_rate": 2.721995147671416e-05,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0045160651206970215,
      "step": 350,
      "valid_targets_mean": 2949.0,
      "valid_targets_min": 305
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.018105922968706292,
      "learning_rate": 2.68032207195547e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003379564033821225,
      "step": 355,
      "valid_targets_mean": 1857.8,
      "valid_targets_min": 341
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.11253667122668588,
      "learning_rate": 2.6383122312019604e-05,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002788994926959276,
      "step": 360,
      "valid_targets_mean": 1678.4,
      "valid_targets_min": 362
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.2819216776790828,
      "learning_rate": 2.595986420628597e-05,
      "loss": 0.0131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002607302973046899,
      "step": 365,
      "valid_targets_mean": 1936.2,
      "valid_targets_min": 454
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.028487037785611482,
      "learning_rate": 2.5533655918607573e-05,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0037446056958287954,
      "step": 370,
      "valid_targets_mean": 2723.1,
      "valid_targets_min": 373
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.14361789153624618,
      "learning_rate": 2.510470842560259e-05,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0025813421234488487,
      "step": 375,
      "valid_targets_mean": 1319.2,
      "valid_targets_min": 381
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.18352715688685126,
      "learning_rate": 2.467323405981841e-05,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0029555675573647022,
      "step": 380,
      "valid_targets_mean": 2413.4,
      "valid_targets_min": 507
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.09834354216900366,
      "learning_rate": 2.423944640462533e-05,
      "loss": 0.0131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003932351246476173,
      "step": 385,
      "valid_targets_mean": 3122.8,
      "valid_targets_min": 351
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.05506373554623298,
      "learning_rate": 2.3803560188490968e-05,
      "loss": 0.0121,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0028243507258594036,
      "step": 390,
      "valid_targets_mean": 2676.0,
      "valid_targets_min": 363
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.10113893234326932,
      "learning_rate": 2.336579117868789e-05,
      "loss": 0.0133,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0036144619807600975,
      "step": 395,
      "valid_targets_mean": 2115.8,
      "valid_targets_min": 349
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.025271752078895756,
      "learning_rate": 2.292635607448711e-05,
      "loss": 0.0134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002341458573937416,
      "step": 400,
      "valid_targets_mean": 1504.8,
      "valid_targets_min": 325
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.1925736365330036,
      "learning_rate": 2.248547239989008e-05,
      "loss": 0.0126,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0026140320114791393,
      "step": 405,
      "valid_targets_mean": 1565.9,
      "valid_targets_min": 391
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.10707482421459247,
      "learning_rate": 2.204335839595255e-05,
      "loss": 0.0126,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003163341199979186,
      "step": 410,
      "valid_targets_mean": 2836.8,
      "valid_targets_min": 347
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.051652169798376495,
      "learning_rate": 2.1600232912753452e-05,
      "loss": 0.0139,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003454884048551321,
      "step": 415,
      "valid_targets_mean": 2927.2,
      "valid_targets_min": 314
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.04623485862462502,
      "learning_rate": 2.1156315301062293e-05,
      "loss": 0.0135,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0030881112907081842,
      "step": 420,
      "valid_targets_mean": 1440.9,
      "valid_targets_min": 378
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.08437804092814283,
      "learning_rate": 2.0711825303758712e-05,
      "loss": 0.0131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004070990718901157,
      "step": 425,
      "valid_targets_mean": 2130.4,
      "valid_targets_min": 474
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.0904753484663695,
      "learning_rate": 2.0266982947057962e-05,
      "loss": 0.0125,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.00331219844520092,
      "step": 430,
      "valid_targets_mean": 3530.5,
      "valid_targets_min": 346
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.15137319759422727,
      "learning_rate": 1.9822008431596083e-05,
      "loss": 0.0135,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003739884588867426,
      "step": 435,
      "valid_targets_mean": 2696.0,
      "valid_targets_min": 438
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.1919119650428584,
      "learning_rate": 1.937712202342881e-05,
      "loss": 0.0131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003298794850707054,
      "step": 440,
      "valid_targets_mean": 1830.1,
      "valid_targets_min": 469
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.24053215880669004,
      "learning_rate": 1.8932543944998037e-05,
      "loss": 0.0133,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0030660992488265038,
      "step": 445,
      "valid_targets_mean": 1606.6,
      "valid_targets_min": 370
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.1306278472074964,
      "learning_rate": 1.8488494266119877e-05,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002405150095000863,
      "step": 450,
      "valid_targets_mean": 2554.1,
      "valid_targets_min": 492
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.28467372784438255,
      "learning_rate": 1.804519279504834e-05,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0037622202653437853,
      "step": 455,
      "valid_targets_mean": 2284.8,
      "valid_targets_min": 445
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.1485547880146562,
      "learning_rate": 1.7602858969668365e-05,
      "loss": 0.0126,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003220830112695694,
      "step": 460,
      "valid_targets_mean": 2098.5,
      "valid_targets_min": 314
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.16461593456165932,
      "learning_rate": 1.716171174887231e-05,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0034352776128798723,
      "step": 465,
      "valid_targets_mean": 2134.6,
      "valid_targets_min": 461
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.03407630963658125,
      "learning_rate": 1.6721969504173484e-05,
      "loss": 0.0137,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002819925546646118,
      "step": 470,
      "valid_targets_mean": 1735.0,
      "valid_targets_min": 318
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.04994478115186678,
      "learning_rate": 1.628384991161041e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003359763417392969,
      "step": 475,
      "valid_targets_mean": 1820.8,
      "valid_targets_min": 392
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.029297557215248998,
      "learning_rate": 1.5847569843995452e-05,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002851391676813364,
      "step": 480,
      "valid_targets_mean": 2505.2,
      "valid_targets_min": 328
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.027402492806578537,
      "learning_rate": 1.5413345263560922e-05,
      "loss": 0.0122,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0030346428975462914,
      "step": 485,
      "valid_targets_mean": 3133.4,
      "valid_targets_min": 392
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.04191942911985486,
      "learning_rate": 1.4981391115056032e-05,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0032312474213540554,
      "step": 490,
      "valid_targets_mean": 2249.8,
      "valid_targets_min": 409
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.04195325222852022,
      "learning_rate": 1.455192121934748e-05,
      "loss": 0.0133,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0036711336579173803,
      "step": 495,
      "valid_targets_mean": 1887.8,
      "valid_targets_min": 322
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.023999859710444302,
      "learning_rate": 1.4125148167576303e-05,
      "loss": 0.0137,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0030063712038099766,
      "step": 500,
      "valid_targets_mean": 1437.1,
      "valid_targets_min": 399
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.11406898270685902,
      "learning_rate": 1.3701283215923563e-05,
      "loss": 0.0134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004196608439087868,
      "step": 505,
      "valid_targets_mean": 3213.6,
      "valid_targets_min": 370
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.1521813456053784,
      "learning_rate": 1.328053618103677e-05,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0027200677432119846,
      "step": 510,
      "valid_targets_mean": 1220.6,
      "valid_targets_min": 367
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.22744695237421958,
      "learning_rate": 1.2863115336168916e-05,
      "loss": 0.013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0031682965345680714,
      "step": 515,
      "valid_targets_mean": 2247.2,
      "valid_targets_min": 189
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.10723504729687629,
      "learning_rate": 1.2449227308081509e-05,
      "loss": 0.0135,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004453069064766169,
      "step": 520,
      "valid_targets_mean": 2142.9,
      "valid_targets_min": 380
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.048289277947422486,
      "learning_rate": 1.2039076974762587e-05,
      "loss": 0.0134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0031748288311064243,
      "step": 525,
      "valid_targets_mean": 1706.8,
      "valid_targets_min": 437
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.019198768255038957,
      "learning_rate": 1.163286736401044e-05,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0033637136220932007,
      "step": 530,
      "valid_targets_mean": 2867.2,
      "valid_targets_min": 516
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.05182505102678586,
      "learning_rate": 1.123079955293322e-05,
      "loss": 0.0125,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003266268875449896,
      "step": 535,
      "valid_targets_mean": 2118.5,
      "valid_targets_min": 448
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.08671391163994449,
      "learning_rate": 1.0833072568414037e-05,
      "loss": 0.0126,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002878316678106785,
      "step": 540,
      "valid_targets_mean": 1528.9,
      "valid_targets_min": 227
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.028144213595349552,
      "learning_rate": 1.0439883288591057e-05,
      "loss": 0.0124,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003090331796556711,
      "step": 545,
      "valid_targets_mean": 2548.9,
      "valid_targets_min": 560
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.01926573012381041,
      "learning_rate": 1.0051426345401202e-05,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035476069897413254,
      "step": 550,
      "valid_targets_mean": 2641.3,
      "valid_targets_min": 388
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.0245042241450859,
      "learning_rate": 9.667894028235704e-06,
      "loss": 0.0127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002284008078277111,
      "step": 555,
      "valid_targets_mean": 1939.3,
      "valid_targets_min": 370
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.06585678679816037,
      "learning_rate": 9.289476188755315e-06,
      "loss": 0.0121,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.00238424027338624,
      "step": 560,
      "valid_targets_mean": 1651.8,
      "valid_targets_min": 375
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.028710945499679674,
      "learning_rate": 8.916360146912122e-06,
      "loss": 0.013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0027281465008854866,
      "step": 565,
      "valid_targets_mean": 1605.7,
      "valid_targets_min": 382
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.033634598074727386,
      "learning_rate": 8.548730598224646e-06,
      "loss": 0.0122,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0025593687314540148,
      "step": 570,
      "valid_targets_mean": 1946.0,
      "valid_targets_min": 316
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.05282113439055737,
      "learning_rate": 8.186769522352053e-06,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0037705537397414446,
      "step": 575,
      "valid_targets_mean": 2199.1,
      "valid_targets_min": 534
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.02998363967591917,
      "learning_rate": 7.830656093012714e-06,
      "loss": 0.0134,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0034412159584462643,
      "step": 580,
      "valid_targets_mean": 2290.3,
      "valid_targets_min": 353
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.028061630837799688,
      "learning_rate": 7.480566589291696e-06,
      "loss": 0.0131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003486450994387269,
      "step": 585,
      "valid_targets_mean": 1681.8,
      "valid_targets_min": 229
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.038326390862229345,
      "learning_rate": 7.1366743083812285e-06,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0037731840275228024,
      "step": 590,
      "valid_targets_mean": 2366.1,
      "valid_targets_min": 514
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.022267188060014886,
      "learning_rate": 6.799149479797101e-06,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0028678697999566793,
      "step": 595,
      "valid_targets_mean": 2442.9,
      "valid_targets_min": 423
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.047417535121547054,
      "learning_rate": 6.4681591811137e-06,
      "loss": 0.0126,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0031579937785863876,
      "step": 600,
      "valid_targets_mean": 2331.9,
      "valid_targets_min": 431
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.041920390871782766,
      "learning_rate": 6.143867255259197e-06,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003032309003174305,
      "step": 605,
      "valid_targets_mean": 2430.2,
      "valid_targets_min": 485
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.03574776156174175,
      "learning_rate": 5.8264342294119504e-06,
      "loss": 0.013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0035699696745723486,
      "step": 610,
      "valid_targets_mean": 2431.1,
      "valid_targets_min": 388
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.03789933951999327,
      "learning_rate": 5.516017235538258e-06,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004032052122056484,
      "step": 615,
      "valid_targets_mean": 3267.8,
      "valid_targets_min": 388
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.05076552580798337,
      "learning_rate": 5.212769932610695e-06,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0021006506867706776,
      "step": 620,
      "valid_targets_mean": 1124.1,
      "valid_targets_min": 347
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.026764161462329343,
      "learning_rate": 4.916842430545681e-06,
      "loss": 0.0127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0032381778582930565,
      "step": 625,
      "valid_targets_mean": 3549.9,
      "valid_targets_min": 360
    },
    {
      "epoch": 4.0128,
      "grad_norm": 0.018080618402527813,
      "learning_rate": 4.628381215897837e-06,
      "loss": 0.0124,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004226190969347954,
      "step": 630,
      "valid_targets_mean": 2953.1,
      "valid_targets_min": 438
    },
    {
      "epoch": 4.0448,
      "grad_norm": 0.04421699271175911,
      "learning_rate": 4.347529079347914e-06,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002999329473823309,
      "step": 635,
      "valid_targets_mean": 1836.9,
      "valid_targets_min": 331
    },
    {
      "epoch": 4.0768,
      "grad_norm": 0.06669320849815788,
      "learning_rate": 4.074425045020247e-06,
      "loss": 0.0125,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0032143457792699337,
      "step": 640,
      "valid_targets_mean": 3027.0,
      "valid_targets_min": 369
    },
    {
      "epoch": 4.1088,
      "grad_norm": 0.025939450064215042,
      "learning_rate": 3.8092043016646487e-06,
      "loss": 0.0132,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002977252472192049,
      "step": 645,
      "valid_targets_mean": 1592.2,
      "valid_targets_min": 378
    },
    {
      "epoch": 4.1408,
      "grad_norm": 0.016682426957388804,
      "learning_rate": 3.551998135736867e-06,
      "loss": 0.0127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003485666122287512,
      "step": 650,
      "valid_targets_mean": 2511.4,
      "valid_targets_min": 447
    },
    {
      "epoch": 4.1728,
      "grad_norm": 0.049555536187823206,
      "learning_rate": 3.3029338664107267e-06,
      "loss": 0.0127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0038854212034493685,
      "step": 655,
      "valid_targets_mean": 3128.3,
      "valid_targets_min": 550
    },
    {
      "epoch": 4.2048,
      "grad_norm": 0.0189134389446547,
      "learning_rate": 3.0621347825540625e-06,
      "loss": 0.0127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0032522277906537056,
      "step": 660,
      "valid_targets_mean": 2686.4,
      "valid_targets_min": 520
    },
    {
      "epoch": 4.2368,
      "grad_norm": 0.031024459905053745,
      "learning_rate": 2.8297200816997183e-06,
      "loss": 0.0126,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0026796162128448486,
      "step": 665,
      "valid_targets_mean": 1200.1,
      "valid_targets_min": 445
    },
    {
      "epoch": 4.2688,
      "grad_norm": 0.02075539090122661,
      "learning_rate": 2.605804811041803e-06,
      "loss": 0.0122,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002532155252993107,
      "step": 670,
      "valid_targets_mean": 1922.6,
      "valid_targets_min": 364
    },
    {
      "epoch": 4.3008,
      "grad_norm": 0.024161319969204228,
      "learning_rate": 2.390499810486351e-06,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002827925141900778,
      "step": 675,
      "valid_targets_mean": 2039.4,
      "valid_targets_min": 378
    },
    {
      "epoch": 4.3328,
      "grad_norm": 0.020720681200900388,
      "learning_rate": 2.183911657784685e-06,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002575571183115244,
      "step": 680,
      "valid_targets_mean": 2230.6,
      "valid_targets_min": 409
    },
    {
      "epoch": 4.3648,
      "grad_norm": 0.02487441263614616,
      "learning_rate": 1.986142615776532e-06,
      "loss": 0.0123,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002919272519648075,
      "step": 685,
      "valid_targets_mean": 2755.1,
      "valid_targets_min": 470
    },
    {
      "epoch": 4.3968,
      "grad_norm": 0.05624312719339281,
      "learning_rate": 1.7972905817690644e-06,
      "loss": 0.0127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003144071437418461,
      "step": 690,
      "valid_targets_mean": 2408.8,
      "valid_targets_min": 381
    },
    {
      "epoch": 4.4288,
      "grad_norm": 0.02919058555403084,
      "learning_rate": 1.617449039076955e-06,
      "loss": 0.013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003166695823892951,
      "step": 695,
      "valid_targets_mean": 1981.8,
      "valid_targets_min": 419
    },
    {
      "epoch": 4.4608,
      "grad_norm": 0.04143469717714482,
      "learning_rate": 1.4467070107473413e-06,
      "loss": 0.0133,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003561312798410654,
      "step": 700,
      "valid_targets_mean": 1753.8,
      "valid_targets_min": 388
    },
    {
      "epoch": 4.4928,
      "grad_norm": 0.03210443926513444,
      "learning_rate": 1.2851490154926816e-06,
      "loss": 0.013,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003960635047405958,
      "step": 705,
      "valid_targets_mean": 2369.0,
      "valid_targets_min": 361
    },
    {
      "epoch": 4.5248,
      "grad_norm": 0.034091344922127405,
      "learning_rate": 1.1328550258533211e-06,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003258173121139407,
      "step": 710,
      "valid_targets_mean": 1886.9,
      "valid_targets_min": 390
    },
    {
      "epoch": 4.5568,
      "grad_norm": 0.0652630794938236,
      "learning_rate": 9.899004286103953e-07,
      "loss": 0.0127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0037811901420354843,
      "step": 715,
      "valid_targets_mean": 2159.1,
      "valid_targets_min": 312
    },
    {
      "epoch": 4.5888,
      "grad_norm": 0.020103370034359653,
      "learning_rate": 8.5635598746876e-07,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0029460175428539515,
      "step": 720,
      "valid_targets_mean": 2170.3,
      "valid_targets_min": 424
    },
    {
      "epoch": 4.6208,
      "grad_norm": 0.021193364258981506,
      "learning_rate": 7.32287808028389e-07,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003635512664914131,
      "step": 725,
      "valid_targets_mean": 2434.9,
      "valid_targets_min": 418
    },
    {
      "epoch": 4.6528,
      "grad_norm": 0.03471844136080961,
      "learning_rate": 6.177573050615327e-07,
      "loss": 0.0131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0031220063101500273,
      "step": 730,
      "valid_targets_mean": 2212.9,
      "valid_targets_min": 353
    },
    {
      "epoch": 4.6848,
      "grad_norm": 0.021504489961340793,
      "learning_rate": 5.128211721119213e-07,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.004023832734674215,
      "step": 735,
      "valid_targets_mean": 2335.6,
      "valid_targets_min": 456
    },
    {
      "epoch": 4.7168,
      "grad_norm": 0.03595855624652171,
      "learning_rate": 4.175313534309755e-07,
      "loss": 0.0127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0026057346258312464,
      "step": 740,
      "valid_targets_mean": 2291.4,
      "valid_targets_min": 375
    },
    {
      "epoch": 4.7488,
      "grad_norm": 0.025445234216983137,
      "learning_rate": 3.319350182649861e-07,
      "loss": 0.0131,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0028861816972494125,
      "step": 745,
      "valid_targets_mean": 1667.3,
      "valid_targets_min": 497
    },
    {
      "epoch": 4.7808,
      "grad_norm": 0.04704153495895216,
      "learning_rate": 2.560745375059392e-07,
      "loss": 0.0125,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.003178564365953207,
      "step": 750,
      "valid_targets_mean": 2123.6,
      "valid_targets_min": 262
    },
    {
      "epoch": 4.8128,
      "grad_norm": 0.021374471295184324,
      "learning_rate": 1.8998746271758016e-07,
      "loss": 0.0122,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002795050386339426,
      "step": 755,
      "valid_targets_mean": 2893.2,
      "valid_targets_min": 389
    },
    {
      "epoch": 4.8448,
      "grad_norm": 0.026756871711829104,
      "learning_rate": 1.337065075470778e-07,
      "loss": 0.0128,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0036452696658670902,
      "step": 760,
      "valid_targets_mean": 2388.3,
      "valid_targets_min": 342
    },
    {
      "epoch": 4.8768,
      "grad_norm": 0.020464291364344217,
      "learning_rate": 8.725953153150279e-08,
      "loss": 0.0129,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0032455329783260822,
      "step": 765,
      "valid_targets_mean": 2036.9,
      "valid_targets_min": 272
    },
    {
      "epoch": 4.9088,
      "grad_norm": 0.028957851696162783,
      "learning_rate": 5.066952630711886e-08,
      "loss": 0.0125,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0030599983874708414,
      "step": 770,
      "valid_targets_mean": 2864.6,
      "valid_targets_min": 408
    },
    {
      "epoch": 4.9408,
      "grad_norm": 0.031287343552416004,
      "learning_rate": 2.3954604228342283e-08,
      "loss": 0.0136,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.002878494095057249,
      "step": 775,
      "valid_targets_mean": 1176.6,
      "valid_targets_min": 331
    },
    {
      "epoch": 4.9728,
      "grad_norm": 0.018333400539361205,
      "learning_rate": 7.12798940197601e-09,
      "loss": 0.0127,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.0022446541115641594,
      "step": 780,
      "valid_targets_mean": 2079.6,
      "valid_targets_min": 463
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.04354659343808137,
      "learning_rate": 1.9801114115480802e-10,
      "loss": 0.0123,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.011615145951509476,
      "step": 785,
      "valid_targets_mean": 2078.6,
      "valid_targets_min": 397
    },
    {
      "epoch": 5.0,
      "loss_nan_ranks": 0,
      "loss_rank_avg": 0.011615145951509476,
      "step": 785,
      "total_flos": 1.0666132853164605e+18,
      "train_loss": 0.014929962234132609,
      "train_runtime": 16979.149,
      "train_samples_per_second": 2.942,
      "train_steps_per_second": 0.046,
      "valid_targets_mean": 2078.6,
      "valid_targets_min": 397
    }
  ],
  "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": 1.0666132853164605e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}