{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 78,
  "global_step": 778,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012857602057216328,
      "grad_norm": 59.91437911987305,
      "learning_rate": 0.0,
      "loss": 2.8221,
      "num_input_tokens_seen": 135296,
      "step": 1,
      "train_runtime": 22.8878,
      "train_tokens_per_second": 5911.283
    },
    {
      "epoch": 0.0025715204114432656,
      "grad_norm": 54.928260803222656,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.8194,
      "num_input_tokens_seen": 295488,
      "step": 2,
      "train_runtime": 38.3163,
      "train_tokens_per_second": 7711.806
    },
    {
      "epoch": 0.003857280617164899,
      "grad_norm": 46.19917297363281,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.5824,
      "num_input_tokens_seen": 390880,
      "step": 3,
      "train_runtime": 46.3319,
      "train_tokens_per_second": 8436.522
    },
    {
      "epoch": 0.005143040822886531,
      "grad_norm": 29.48894691467285,
      "learning_rate": 1.2e-05,
      "loss": 2.1127,
      "num_input_tokens_seen": 527584,
      "step": 4,
      "train_runtime": 58.9913,
      "train_tokens_per_second": 8943.425
    },
    {
      "epoch": 0.006428801028608165,
      "grad_norm": 12.14394474029541,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.7084,
      "num_input_tokens_seen": 644736,
      "step": 5,
      "train_runtime": 69.2931,
      "train_tokens_per_second": 9304.479
    },
    {
      "epoch": 0.007714561234329798,
      "grad_norm": 9.809494018554688,
      "learning_rate": 2e-05,
      "loss": 1.681,
      "num_input_tokens_seen": 777056,
      "step": 6,
      "train_runtime": 81.2029,
      "train_tokens_per_second": 9569.309
    },
    {
      "epoch": 0.00900032144005143,
      "grad_norm": 8.21720027923584,
      "learning_rate": 1.997412677878396e-05,
      "loss": 1.6744,
      "num_input_tokens_seen": 910560,
      "step": 7,
      "train_runtime": 93.321,
      "train_tokens_per_second": 9757.29
    },
    {
      "epoch": 0.010286081645773062,
      "grad_norm": 7.667925834655762,
      "learning_rate": 1.994825355756792e-05,
      "loss": 1.4179,
      "num_input_tokens_seen": 1062912,
      "step": 8,
      "train_runtime": 107.311,
      "train_tokens_per_second": 9904.967
    },
    {
      "epoch": 0.011571841851494697,
      "grad_norm": 6.243086814880371,
      "learning_rate": 1.9922380336351877e-05,
      "loss": 1.3252,
      "num_input_tokens_seen": 1152448,
      "step": 9,
      "train_runtime": 114.8712,
      "train_tokens_per_second": 10032.519
    },
    {
      "epoch": 0.01285760205721633,
      "grad_norm": 4.096212863922119,
      "learning_rate": 1.9896507115135837e-05,
      "loss": 1.2179,
      "num_input_tokens_seen": 1251424,
      "step": 10,
      "train_runtime": 123.394,
      "train_tokens_per_second": 10141.69
    },
    {
      "epoch": 0.014143362262937963,
      "grad_norm": 3.7918293476104736,
      "learning_rate": 1.9870633893919795e-05,
      "loss": 1.3145,
      "num_input_tokens_seen": 1401632,
      "step": 11,
      "train_runtime": 137.3677,
      "train_tokens_per_second": 10203.506
    },
    {
      "epoch": 0.015429122468659595,
      "grad_norm": 2.6636881828308105,
      "learning_rate": 1.9844760672703752e-05,
      "loss": 1.1167,
      "num_input_tokens_seen": 1485088,
      "step": 12,
      "train_runtime": 144.3866,
      "train_tokens_per_second": 10285.495
    },
    {
      "epoch": 0.01671488267438123,
      "grad_norm": 3.1536850929260254,
      "learning_rate": 1.9818887451487713e-05,
      "loss": 1.2975,
      "num_input_tokens_seen": 1599424,
      "step": 13,
      "train_runtime": 154.4368,
      "train_tokens_per_second": 10356.492
    },
    {
      "epoch": 0.01800064288010286,
      "grad_norm": 2.806964874267578,
      "learning_rate": 1.979301423027167e-05,
      "loss": 1.2166,
      "num_input_tokens_seen": 1688192,
      "step": 14,
      "train_runtime": 161.9986,
      "train_tokens_per_second": 10421.031
    },
    {
      "epoch": 0.019286403085824494,
      "grad_norm": 2.8832125663757324,
      "learning_rate": 1.9767141009055627e-05,
      "loss": 1.3103,
      "num_input_tokens_seen": 1779040,
      "step": 15,
      "train_runtime": 169.8234,
      "train_tokens_per_second": 10475.825
    },
    {
      "epoch": 0.020572163291546125,
      "grad_norm": 2.3326032161712646,
      "learning_rate": 1.9741267787839588e-05,
      "loss": 1.1936,
      "num_input_tokens_seen": 1898976,
      "step": 16,
      "train_runtime": 180.4047,
      "train_tokens_per_second": 10526.202
    },
    {
      "epoch": 0.02185792349726776,
      "grad_norm": 2.495771646499634,
      "learning_rate": 1.971539456662355e-05,
      "loss": 1.238,
      "num_input_tokens_seen": 2002496,
      "step": 17,
      "train_runtime": 189.3388,
      "train_tokens_per_second": 10576.258
    },
    {
      "epoch": 0.023143683702989394,
      "grad_norm": 2.3127689361572266,
      "learning_rate": 1.9689521345407506e-05,
      "loss": 1.2386,
      "num_input_tokens_seen": 2148032,
      "step": 18,
      "train_runtime": 202.5493,
      "train_tokens_per_second": 10604.983
    },
    {
      "epoch": 0.024429443908711025,
      "grad_norm": 2.1507444381713867,
      "learning_rate": 1.9663648124191463e-05,
      "loss": 1.1955,
      "num_input_tokens_seen": 2273632,
      "step": 19,
      "train_runtime": 214.0022,
      "train_tokens_per_second": 10624.337
    },
    {
      "epoch": 0.02571520411443266,
      "grad_norm": 2.7235100269317627,
      "learning_rate": 1.963777490297542e-05,
      "loss": 1.2066,
      "num_input_tokens_seen": 2383520,
      "step": 20,
      "train_runtime": 223.8834,
      "train_tokens_per_second": 10646.255
    },
    {
      "epoch": 0.02700096432015429,
      "grad_norm": 2.299577236175537,
      "learning_rate": 1.961190168175938e-05,
      "loss": 1.1685,
      "num_input_tokens_seen": 2483456,
      "step": 21,
      "train_runtime": 232.4787,
      "train_tokens_per_second": 10682.509
    },
    {
      "epoch": 0.028286724525875925,
      "grad_norm": 2.1814215183258057,
      "learning_rate": 1.9586028460543338e-05,
      "loss": 1.1816,
      "num_input_tokens_seen": 2628480,
      "step": 22,
      "train_runtime": 245.7507,
      "train_tokens_per_second": 10695.718
    },
    {
      "epoch": 0.029572484731597556,
      "grad_norm": 1.9551724195480347,
      "learning_rate": 1.95601552393273e-05,
      "loss": 1.1728,
      "num_input_tokens_seen": 2741984,
      "step": 23,
      "train_runtime": 255.7533,
      "train_tokens_per_second": 10721.207
    },
    {
      "epoch": 0.03085824493731919,
      "grad_norm": 1.9535983800888062,
      "learning_rate": 1.9534282018111256e-05,
      "loss": 1.2263,
      "num_input_tokens_seen": 2827744,
      "step": 24,
      "train_runtime": 263.0783,
      "train_tokens_per_second": 10748.679
    },
    {
      "epoch": 0.032144005143040826,
      "grad_norm": 2.2656548023223877,
      "learning_rate": 1.9508408796895217e-05,
      "loss": 1.1648,
      "num_input_tokens_seen": 2926304,
      "step": 25,
      "train_runtime": 271.5983,
      "train_tokens_per_second": 10774.384
    },
    {
      "epoch": 0.03342976534876246,
      "grad_norm": 2.323800563812256,
      "learning_rate": 1.9482535575679174e-05,
      "loss": 1.1831,
      "num_input_tokens_seen": 3039168,
      "step": 26,
      "train_runtime": 281.843,
      "train_tokens_per_second": 10783.195
    },
    {
      "epoch": 0.03471552555448409,
      "grad_norm": 1.9560903310775757,
      "learning_rate": 1.945666235446313e-05,
      "loss": 1.2156,
      "num_input_tokens_seen": 3162080,
      "step": 27,
      "train_runtime": 293.1326,
      "train_tokens_per_second": 10787.199
    },
    {
      "epoch": 0.03600128576020572,
      "grad_norm": 1.7636638879776,
      "learning_rate": 1.943078913324709e-05,
      "loss": 1.1564,
      "num_input_tokens_seen": 3277408,
      "step": 28,
      "train_runtime": 303.3144,
      "train_tokens_per_second": 10805.316
    },
    {
      "epoch": 0.03728704596592736,
      "grad_norm": 1.9402304887771606,
      "learning_rate": 1.940491591203105e-05,
      "loss": 1.2378,
      "num_input_tokens_seen": 3393408,
      "step": 29,
      "train_runtime": 313.5538,
      "train_tokens_per_second": 10822.41
    },
    {
      "epoch": 0.03857280617164899,
      "grad_norm": 2.0136935710906982,
      "learning_rate": 1.937904269081501e-05,
      "loss": 1.2205,
      "num_input_tokens_seen": 3493664,
      "step": 30,
      "train_runtime": 322.3163,
      "train_tokens_per_second": 10839.242
    },
    {
      "epoch": 0.03985856637737062,
      "grad_norm": 2.0718331336975098,
      "learning_rate": 1.9353169469598967e-05,
      "loss": 1.2168,
      "num_input_tokens_seen": 3632128,
      "step": 31,
      "train_runtime": 351.6221,
      "train_tokens_per_second": 10329.636
    },
    {
      "epoch": 0.04114432658309225,
      "grad_norm": 1.8790080547332764,
      "learning_rate": 1.9327296248382924e-05,
      "loss": 1.1357,
      "num_input_tokens_seen": 3740256,
      "step": 32,
      "train_runtime": 360.9507,
      "train_tokens_per_second": 10362.236
    },
    {
      "epoch": 0.04243008678881389,
      "grad_norm": 1.775802731513977,
      "learning_rate": 1.9301423027166882e-05,
      "loss": 1.1974,
      "num_input_tokens_seen": 3836320,
      "step": 33,
      "train_runtime": 369.242,
      "train_tokens_per_second": 10389.716
    },
    {
      "epoch": 0.04371584699453552,
      "grad_norm": 1.7465744018554688,
      "learning_rate": 1.9275549805950842e-05,
      "loss": 1.1552,
      "num_input_tokens_seen": 3973216,
      "step": 34,
      "train_runtime": 381.9458,
      "train_tokens_per_second": 10402.566
    },
    {
      "epoch": 0.04500160720025715,
      "grad_norm": 1.946832299232483,
      "learning_rate": 1.9249676584734803e-05,
      "loss": 1.0835,
      "num_input_tokens_seen": 4098496,
      "step": 35,
      "train_runtime": 393.2749,
      "train_tokens_per_second": 10421.453
    },
    {
      "epoch": 0.04628736740597879,
      "grad_norm": 1.8845430612564087,
      "learning_rate": 1.922380336351876e-05,
      "loss": 1.2099,
      "num_input_tokens_seen": 4211328,
      "step": 36,
      "train_runtime": 403.0748,
      "train_tokens_per_second": 10448.006
    },
    {
      "epoch": 0.04757312761170042,
      "grad_norm": 1.9792522192001343,
      "learning_rate": 1.9197930142302718e-05,
      "loss": 1.1743,
      "num_input_tokens_seen": 4354368,
      "step": 37,
      "train_runtime": 416.1989,
      "train_tokens_per_second": 10462.229
    },
    {
      "epoch": 0.04885888781742205,
      "grad_norm": 1.7486963272094727,
      "learning_rate": 1.9172056921086678e-05,
      "loss": 1.1916,
      "num_input_tokens_seen": 4459872,
      "step": 38,
      "train_runtime": 425.2452,
      "train_tokens_per_second": 10487.765
    },
    {
      "epoch": 0.05014464802314368,
      "grad_norm": 1.8029528856277466,
      "learning_rate": 1.9146183699870636e-05,
      "loss": 1.2058,
      "num_input_tokens_seen": 4586752,
      "step": 39,
      "train_runtime": 436.5305,
      "train_tokens_per_second": 10507.289
    },
    {
      "epoch": 0.05143040822886532,
      "grad_norm": 1.6773114204406738,
      "learning_rate": 1.9120310478654593e-05,
      "loss": 1.171,
      "num_input_tokens_seen": 4679744,
      "step": 40,
      "train_runtime": 444.5403,
      "train_tokens_per_second": 10527.152
    },
    {
      "epoch": 0.05271616843458695,
      "grad_norm": 2.0230824947357178,
      "learning_rate": 1.909443725743855e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 4834496,
      "step": 41,
      "train_runtime": 459.2262,
      "train_tokens_per_second": 10527.482
    },
    {
      "epoch": 0.05400192864030858,
      "grad_norm": 1.7988804578781128,
      "learning_rate": 1.906856403622251e-05,
      "loss": 1.1448,
      "num_input_tokens_seen": 4924224,
      "step": 42,
      "train_runtime": 466.8323,
      "train_tokens_per_second": 10548.165
    },
    {
      "epoch": 0.05528768884603021,
      "grad_norm": 1.8624871969223022,
      "learning_rate": 1.904269081500647e-05,
      "loss": 1.0607,
      "num_input_tokens_seen": 5027584,
      "step": 43,
      "train_runtime": 475.733,
      "train_tokens_per_second": 10568.078
    },
    {
      "epoch": 0.05657344905175185,
      "grad_norm": 1.902448296546936,
      "learning_rate": 1.901681759379043e-05,
      "loss": 1.2209,
      "num_input_tokens_seen": 5147008,
      "step": 44,
      "train_runtime": 486.1981,
      "train_tokens_per_second": 10586.237
    },
    {
      "epoch": 0.05785920925747348,
      "grad_norm": 1.875640869140625,
      "learning_rate": 1.8990944372574386e-05,
      "loss": 1.1073,
      "num_input_tokens_seen": 5265504,
      "step": 45,
      "train_runtime": 497.0036,
      "train_tokens_per_second": 10594.498
    },
    {
      "epoch": 0.05914496946319511,
      "grad_norm": 1.8521565198898315,
      "learning_rate": 1.8965071151358347e-05,
      "loss": 1.1615,
      "num_input_tokens_seen": 5367456,
      "step": 46,
      "train_runtime": 505.7756,
      "train_tokens_per_second": 10612.328
    },
    {
      "epoch": 0.060430729668916744,
      "grad_norm": 1.8101763725280762,
      "learning_rate": 1.8939197930142304e-05,
      "loss": 1.0999,
      "num_input_tokens_seen": 5491136,
      "step": 47,
      "train_runtime": 516.6042,
      "train_tokens_per_second": 10629.291
    },
    {
      "epoch": 0.06171648987463838,
      "grad_norm": 1.81596040725708,
      "learning_rate": 1.8913324708926265e-05,
      "loss": 1.0616,
      "num_input_tokens_seen": 5608224,
      "step": 48,
      "train_runtime": 527.1638,
      "train_tokens_per_second": 10638.484
    },
    {
      "epoch": 0.06300225008036,
      "grad_norm": 1.988157868385315,
      "learning_rate": 1.8887451487710222e-05,
      "loss": 1.2014,
      "num_input_tokens_seen": 5729248,
      "step": 49,
      "train_runtime": 538.0492,
      "train_tokens_per_second": 10648.187
    },
    {
      "epoch": 0.06428801028608165,
      "grad_norm": 1.8147128820419312,
      "learning_rate": 1.886157826649418e-05,
      "loss": 1.0791,
      "num_input_tokens_seen": 5888672,
      "step": 50,
      "train_runtime": 552.8286,
      "train_tokens_per_second": 10651.895
    },
    {
      "epoch": 0.06557377049180328,
      "grad_norm": 1.8205593824386597,
      "learning_rate": 1.883570504527814e-05,
      "loss": 1.1579,
      "num_input_tokens_seen": 6022656,
      "step": 51,
      "train_runtime": 565.0367,
      "train_tokens_per_second": 10658.875
    },
    {
      "epoch": 0.06685953069752491,
      "grad_norm": 1.8702155351638794,
      "learning_rate": 1.8809831824062097e-05,
      "loss": 1.1736,
      "num_input_tokens_seen": 6122848,
      "step": 52,
      "train_runtime": 573.6417,
      "train_tokens_per_second": 10673.646
    },
    {
      "epoch": 0.06814529090324654,
      "grad_norm": 2.053255796432495,
      "learning_rate": 1.8783958602846054e-05,
      "loss": 1.1884,
      "num_input_tokens_seen": 6219968,
      "step": 53,
      "train_runtime": 581.9801,
      "train_tokens_per_second": 10687.596
    },
    {
      "epoch": 0.06943105110896818,
      "grad_norm": 1.8326001167297363,
      "learning_rate": 1.8758085381630015e-05,
      "loss": 1.1875,
      "num_input_tokens_seen": 6330400,
      "step": 54,
      "train_runtime": 591.5047,
      "train_tokens_per_second": 10702.198
    },
    {
      "epoch": 0.0707168113146898,
      "grad_norm": 1.9650863409042358,
      "learning_rate": 1.8732212160413976e-05,
      "loss": 1.1747,
      "num_input_tokens_seen": 6500608,
      "step": 55,
      "train_runtime": 607.6429,
      "train_tokens_per_second": 10698.072
    },
    {
      "epoch": 0.07200257152041144,
      "grad_norm": 1.8934389352798462,
      "learning_rate": 1.8706338939197933e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 6600288,
      "step": 56,
      "train_runtime": 616.1735,
      "train_tokens_per_second": 10711.735
    },
    {
      "epoch": 0.07328833172613308,
      "grad_norm": 1.8143988847732544,
      "learning_rate": 1.868046571798189e-05,
      "loss": 1.1152,
      "num_input_tokens_seen": 6713184,
      "step": 57,
      "train_runtime": 625.8716,
      "train_tokens_per_second": 10726.135
    },
    {
      "epoch": 0.07457409193185471,
      "grad_norm": 1.8745934963226318,
      "learning_rate": 1.8654592496765847e-05,
      "loss": 1.1131,
      "num_input_tokens_seen": 6854400,
      "step": 58,
      "train_runtime": 638.8924,
      "train_tokens_per_second": 10728.567
    },
    {
      "epoch": 0.07585985213757634,
      "grad_norm": 1.8524185419082642,
      "learning_rate": 1.8628719275549808e-05,
      "loss": 1.0552,
      "num_input_tokens_seen": 6951744,
      "step": 59,
      "train_runtime": 647.376,
      "train_tokens_per_second": 10738.341
    },
    {
      "epoch": 0.07714561234329798,
      "grad_norm": 1.8481533527374268,
      "learning_rate": 1.8602846054333765e-05,
      "loss": 1.1928,
      "num_input_tokens_seen": 7081184,
      "step": 60,
      "train_runtime": 659.3929,
      "train_tokens_per_second": 10738.945
    },
    {
      "epoch": 0.0784313725490196,
      "grad_norm": 1.8214608430862427,
      "learning_rate": 1.8576972833117726e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 7226528,
      "step": 61,
      "train_runtime": 688.7221,
      "train_tokens_per_second": 10492.661
    },
    {
      "epoch": 0.07971713275474124,
      "grad_norm": 1.8268235921859741,
      "learning_rate": 1.8551099611901683e-05,
      "loss": 1.0817,
      "num_input_tokens_seen": 7317344,
      "step": 62,
      "train_runtime": 696.3751,
      "train_tokens_per_second": 10507.762
    },
    {
      "epoch": 0.08100289296046287,
      "grad_norm": 1.8014057874679565,
      "learning_rate": 1.8525226390685644e-05,
      "loss": 1.1799,
      "num_input_tokens_seen": 7473824,
      "step": 63,
      "train_runtime": 711.0738,
      "train_tokens_per_second": 10510.616
    },
    {
      "epoch": 0.0822886531661845,
      "grad_norm": 1.826602816581726,
      "learning_rate": 1.84993531694696e-05,
      "loss": 1.1546,
      "num_input_tokens_seen": 7575392,
      "step": 64,
      "train_runtime": 719.8185,
      "train_tokens_per_second": 10524.031
    },
    {
      "epoch": 0.08357441337190614,
      "grad_norm": 1.7211124897003174,
      "learning_rate": 1.847347994825356e-05,
      "loss": 1.1899,
      "num_input_tokens_seen": 7723680,
      "step": 65,
      "train_runtime": 733.4703,
      "train_tokens_per_second": 10530.324
    },
    {
      "epoch": 0.08486017357762778,
      "grad_norm": 1.7656164169311523,
      "learning_rate": 1.8447606727037516e-05,
      "loss": 1.2178,
      "num_input_tokens_seen": 7850144,
      "step": 66,
      "train_runtime": 745.1189,
      "train_tokens_per_second": 10535.425
    },
    {
      "epoch": 0.08614593378334941,
      "grad_norm": 1.8083853721618652,
      "learning_rate": 1.8421733505821476e-05,
      "loss": 1.1542,
      "num_input_tokens_seen": 7950752,
      "step": 67,
      "train_runtime": 753.8917,
      "train_tokens_per_second": 10546.278
    },
    {
      "epoch": 0.08743169398907104,
      "grad_norm": 1.7961894273757935,
      "learning_rate": 1.8395860284605437e-05,
      "loss": 1.044,
      "num_input_tokens_seen": 8080160,
      "step": 68,
      "train_runtime": 765.8094,
      "train_tokens_per_second": 10551.138
    },
    {
      "epoch": 0.08871745419479267,
      "grad_norm": 1.994905710220337,
      "learning_rate": 1.8369987063389394e-05,
      "loss": 1.1235,
      "num_input_tokens_seen": 8218784,
      "step": 69,
      "train_runtime": 778.4148,
      "train_tokens_per_second": 10558.36
    },
    {
      "epoch": 0.0900032144005143,
      "grad_norm": 1.9451382160186768,
      "learning_rate": 1.834411384217335e-05,
      "loss": 1.1517,
      "num_input_tokens_seen": 8333600,
      "step": 70,
      "train_runtime": 788.6364,
      "train_tokens_per_second": 10567.1
    },
    {
      "epoch": 0.09128897460623593,
      "grad_norm": 1.8312456607818604,
      "learning_rate": 1.8318240620957312e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 8460128,
      "step": 71,
      "train_runtime": 799.7803,
      "train_tokens_per_second": 10578.065
    },
    {
      "epoch": 0.09257473481195758,
      "grad_norm": 1.9686640501022339,
      "learning_rate": 1.829236739974127e-05,
      "loss": 1.2261,
      "num_input_tokens_seen": 8536256,
      "step": 72,
      "train_runtime": 806.2621,
      "train_tokens_per_second": 10587.446
    },
    {
      "epoch": 0.09386049501767921,
      "grad_norm": 1.8814424276351929,
      "learning_rate": 1.8266494178525227e-05,
      "loss": 1.0726,
      "num_input_tokens_seen": 8641280,
      "step": 73,
      "train_runtime": 815.3389,
      "train_tokens_per_second": 10598.39
    },
    {
      "epoch": 0.09514625522340084,
      "grad_norm": 1.9460242986679077,
      "learning_rate": 1.8240620957309187e-05,
      "loss": 1.1726,
      "num_input_tokens_seen": 8743456,
      "step": 74,
      "train_runtime": 824.1743,
      "train_tokens_per_second": 10608.746
    },
    {
      "epoch": 0.09643201542912247,
      "grad_norm": 2.0222864151000977,
      "learning_rate": 1.8214747736093145e-05,
      "loss": 1.1902,
      "num_input_tokens_seen": 8850720,
      "step": 75,
      "train_runtime": 833.6218,
      "train_tokens_per_second": 10617.189
    },
    {
      "epoch": 0.0977177756348441,
      "grad_norm": 1.943700909614563,
      "learning_rate": 1.8188874514877105e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 8990112,
      "step": 76,
      "train_runtime": 846.3567,
      "train_tokens_per_second": 10622.131
    },
    {
      "epoch": 0.09900353584056573,
      "grad_norm": 2.1447455883026123,
      "learning_rate": 1.8163001293661063e-05,
      "loss": 1.1458,
      "num_input_tokens_seen": 9074080,
      "step": 77,
      "train_runtime": 853.5667,
      "train_tokens_per_second": 10630.78
    },
    {
      "epoch": 0.10028929604628736,
      "grad_norm": 2.0305962562561035,
      "learning_rate": 1.813712807244502e-05,
      "loss": 1.0369,
      "num_input_tokens_seen": 9180160,
      "step": 78,
      "train_runtime": 862.9817,
      "train_tokens_per_second": 10637.723
    },
    {
      "epoch": 0.10028929604628736,
      "eval_loss": 1.1576727628707886,
      "eval_runtime": 23.0724,
      "eval_samples_per_second": 43.082,
      "eval_steps_per_second": 1.387,
      "num_input_tokens_seen": 9180160,
      "step": 78
    },
    {
      "epoch": 0.101575056252009,
      "grad_norm": 1.998412847518921,
      "learning_rate": 1.8111254851228977e-05,
      "loss": 1.145,
      "num_input_tokens_seen": 9270560,
      "step": 79,
      "train_runtime": 893.8501,
      "train_tokens_per_second": 10371.492
    },
    {
      "epoch": 0.10286081645773064,
      "grad_norm": 1.8889389038085938,
      "learning_rate": 1.8085381630012938e-05,
      "loss": 1.1677,
      "num_input_tokens_seen": 9389504,
      "step": 80,
      "train_runtime": 904.2477,
      "train_tokens_per_second": 10383.774
    },
    {
      "epoch": 0.10414657666345227,
      "grad_norm": 1.8141217231750488,
      "learning_rate": 1.80595084087969e-05,
      "loss": 1.0648,
      "num_input_tokens_seen": 9498464,
      "step": 81,
      "train_runtime": 913.7098,
      "train_tokens_per_second": 10395.494
    },
    {
      "epoch": 0.1054323368691739,
      "grad_norm": 1.8328213691711426,
      "learning_rate": 1.8033635187580856e-05,
      "loss": 1.1208,
      "num_input_tokens_seen": 9619936,
      "step": 82,
      "train_runtime": 924.4896,
      "train_tokens_per_second": 10405.672
    },
    {
      "epoch": 0.10671809707489553,
      "grad_norm": 1.795311689376831,
      "learning_rate": 1.8007761966364813e-05,
      "loss": 1.0181,
      "num_input_tokens_seen": 9726976,
      "step": 83,
      "train_runtime": 933.8495,
      "train_tokens_per_second": 10416.0
    },
    {
      "epoch": 0.10800385728061716,
      "grad_norm": 1.8454675674438477,
      "learning_rate": 1.7981888745148774e-05,
      "loss": 1.0702,
      "num_input_tokens_seen": 9828896,
      "step": 84,
      "train_runtime": 942.6626,
      "train_tokens_per_second": 10426.738
    },
    {
      "epoch": 0.1092896174863388,
      "grad_norm": 1.9122002124786377,
      "learning_rate": 1.795601552393273e-05,
      "loss": 1.184,
      "num_input_tokens_seen": 10007680,
      "step": 85,
      "train_runtime": 960.0623,
      "train_tokens_per_second": 10423.99
    },
    {
      "epoch": 0.11057537769206043,
      "grad_norm": 1.873637318611145,
      "learning_rate": 1.7930142302716688e-05,
      "loss": 1.1323,
      "num_input_tokens_seen": 10161088,
      "step": 86,
      "train_runtime": 974.6068,
      "train_tokens_per_second": 10425.833
    },
    {
      "epoch": 0.11186113789778207,
      "grad_norm": 1.7707033157348633,
      "learning_rate": 1.790426908150065e-05,
      "loss": 1.0968,
      "num_input_tokens_seen": 10247072,
      "step": 87,
      "train_runtime": 981.9572,
      "train_tokens_per_second": 10435.355
    },
    {
      "epoch": 0.1131468981035037,
      "grad_norm": 1.7864853143692017,
      "learning_rate": 1.7878395860284606e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 10438112,
      "step": 88,
      "train_runtime": 1000.9105,
      "train_tokens_per_second": 10428.617
    },
    {
      "epoch": 0.11443265830922533,
      "grad_norm": 1.8133771419525146,
      "learning_rate": 1.7852522639068567e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 10559456,
      "step": 89,
      "train_runtime": 1012.136,
      "train_tokens_per_second": 10432.843
    },
    {
      "epoch": 0.11571841851494696,
      "grad_norm": 1.854591965675354,
      "learning_rate": 1.7826649417852524e-05,
      "loss": 1.1079,
      "num_input_tokens_seen": 10675840,
      "step": 90,
      "train_runtime": 1022.5885,
      "train_tokens_per_second": 10440.015
    },
    {
      "epoch": 0.1170041787206686,
      "grad_norm": 1.7617478370666504,
      "learning_rate": 1.780077619663648e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 10791936,
      "step": 91,
      "train_runtime": 1048.7392,
      "train_tokens_per_second": 10290.39
    },
    {
      "epoch": 0.11828993892639023,
      "grad_norm": 1.9446653127670288,
      "learning_rate": 1.7774902975420442e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 10917536,
      "step": 92,
      "train_runtime": 1059.9578,
      "train_tokens_per_second": 10299.972
    },
    {
      "epoch": 0.11957569913211186,
      "grad_norm": 1.8482822179794312,
      "learning_rate": 1.7749029754204403e-05,
      "loss": 1.1365,
      "num_input_tokens_seen": 11021824,
      "step": 93,
      "train_runtime": 1069.2663,
      "train_tokens_per_second": 10307.838
    },
    {
      "epoch": 0.12086145933783349,
      "grad_norm": 1.8727803230285645,
      "learning_rate": 1.772315653298836e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 11159872,
      "step": 94,
      "train_runtime": 1081.6899,
      "train_tokens_per_second": 10317.071
    },
    {
      "epoch": 0.12214721954355513,
      "grad_norm": 1.779184341430664,
      "learning_rate": 1.7697283311772317e-05,
      "loss": 1.0668,
      "num_input_tokens_seen": 11258336,
      "step": 95,
      "train_runtime": 1090.2473,
      "train_tokens_per_second": 10326.406
    },
    {
      "epoch": 0.12343297974927676,
      "grad_norm": 1.8586459159851074,
      "learning_rate": 1.7671410090556274e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 11381312,
      "step": 96,
      "train_runtime": 1101.6291,
      "train_tokens_per_second": 10331.347
    },
    {
      "epoch": 0.1247187399549984,
      "grad_norm": 1.8665542602539062,
      "learning_rate": 1.7645536869340235e-05,
      "loss": 1.108,
      "num_input_tokens_seen": 11554048,
      "step": 97,
      "train_runtime": 1118.3041,
      "train_tokens_per_second": 10331.759
    },
    {
      "epoch": 0.12600450016072,
      "grad_norm": 1.8169111013412476,
      "learning_rate": 1.7619663648124192e-05,
      "loss": 1.1525,
      "num_input_tokens_seen": 11717152,
      "step": 98,
      "train_runtime": 1134.1337,
      "train_tokens_per_second": 10331.367
    },
    {
      "epoch": 0.12729026036644167,
      "grad_norm": 1.9255194664001465,
      "learning_rate": 1.759379042690815e-05,
      "loss": 1.0878,
      "num_input_tokens_seen": 11854432,
      "step": 99,
      "train_runtime": 1146.8016,
      "train_tokens_per_second": 10336.951
    },
    {
      "epoch": 0.1285760205721633,
      "grad_norm": 1.9415030479431152,
      "learning_rate": 1.756791720569211e-05,
      "loss": 1.1299,
      "num_input_tokens_seen": 11979232,
      "step": 100,
      "train_runtime": 1158.3947,
      "train_tokens_per_second": 10341.235
    },
    {
      "epoch": 0.12986178077788493,
      "grad_norm": 1.9196245670318604,
      "learning_rate": 1.754204398447607e-05,
      "loss": 1.124,
      "num_input_tokens_seen": 12144416,
      "step": 101,
      "train_runtime": 1173.9193,
      "train_tokens_per_second": 10345.188
    },
    {
      "epoch": 0.13114754098360656,
      "grad_norm": 1.8218384981155396,
      "learning_rate": 1.7516170763260028e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 12261408,
      "step": 102,
      "train_runtime": 1184.2447,
      "train_tokens_per_second": 10353.779
    },
    {
      "epoch": 0.1324333011893282,
      "grad_norm": 1.86562979221344,
      "learning_rate": 1.7490297542043985e-05,
      "loss": 1.0873,
      "num_input_tokens_seen": 12384448,
      "step": 103,
      "train_runtime": 1194.8569,
      "train_tokens_per_second": 10364.796
    },
    {
      "epoch": 0.13371906139504983,
      "grad_norm": 1.9434869289398193,
      "learning_rate": 1.7464424320827943e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 12476352,
      "step": 104,
      "train_runtime": 1202.6401,
      "train_tokens_per_second": 10374.136
    },
    {
      "epoch": 0.13500482160077146,
      "grad_norm": 1.9082143306732178,
      "learning_rate": 1.7438551099611903e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 12562464,
      "step": 105,
      "train_runtime": 1209.8616,
      "train_tokens_per_second": 10383.39
    },
    {
      "epoch": 0.1362905818064931,
      "grad_norm": 1.7863415479660034,
      "learning_rate": 1.7412677878395864e-05,
      "loss": 1.1061,
      "num_input_tokens_seen": 12654304,
      "step": 106,
      "train_runtime": 1217.5491,
      "train_tokens_per_second": 10393.26
    },
    {
      "epoch": 0.13757634201221472,
      "grad_norm": 1.872205138206482,
      "learning_rate": 1.738680465717982e-05,
      "loss": 0.9919,
      "num_input_tokens_seen": 12782144,
      "step": 107,
      "train_runtime": 1229.0168,
      "train_tokens_per_second": 10400.3
    },
    {
      "epoch": 0.13886210221793635,
      "grad_norm": 2.2330684661865234,
      "learning_rate": 1.736093143596378e-05,
      "loss": 1.1528,
      "num_input_tokens_seen": 12899840,
      "step": 108,
      "train_runtime": 1239.2467,
      "train_tokens_per_second": 10409.421
    },
    {
      "epoch": 0.14014786242365798,
      "grad_norm": 2.096357822418213,
      "learning_rate": 1.733505821474774e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 13043392,
      "step": 109,
      "train_runtime": 1252.1584,
      "train_tokens_per_second": 10416.727
    },
    {
      "epoch": 0.1414336226293796,
      "grad_norm": 1.9565867185592651,
      "learning_rate": 1.7309184993531696e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 13188448,
      "step": 110,
      "train_runtime": 1265.437,
      "train_tokens_per_second": 10422.05
    },
    {
      "epoch": 0.14271938283510124,
      "grad_norm": 1.9779455661773682,
      "learning_rate": 1.7283311772315654e-05,
      "loss": 1.0885,
      "num_input_tokens_seen": 13275616,
      "step": 111,
      "train_runtime": 1272.8268,
      "train_tokens_per_second": 10430.026
    },
    {
      "epoch": 0.14400514304082287,
      "grad_norm": 1.8848296403884888,
      "learning_rate": 1.725743855109961e-05,
      "loss": 1.0563,
      "num_input_tokens_seen": 13401984,
      "step": 112,
      "train_runtime": 1284.164,
      "train_tokens_per_second": 10436.35
    },
    {
      "epoch": 0.1452909032465445,
      "grad_norm": 1.8709964752197266,
      "learning_rate": 1.723156532988357e-05,
      "loss": 1.1049,
      "num_input_tokens_seen": 13516096,
      "step": 113,
      "train_runtime": 1294.3698,
      "train_tokens_per_second": 10442.221
    },
    {
      "epoch": 0.14657666345226616,
      "grad_norm": 1.7715637683868408,
      "learning_rate": 1.7205692108667532e-05,
      "loss": 1.0666,
      "num_input_tokens_seen": 13608128,
      "step": 114,
      "train_runtime": 1302.3125,
      "train_tokens_per_second": 10449.203
    },
    {
      "epoch": 0.1478624236579878,
      "grad_norm": 1.7707303762435913,
      "learning_rate": 1.717981888745149e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 13708288,
      "step": 115,
      "train_runtime": 1310.8819,
      "train_tokens_per_second": 10457.302
    },
    {
      "epoch": 0.14914818386370943,
      "grad_norm": 2.0213918685913086,
      "learning_rate": 1.7153945666235447e-05,
      "loss": 1.1718,
      "num_input_tokens_seen": 13832640,
      "step": 116,
      "train_runtime": 1321.8066,
      "train_tokens_per_second": 10464.95
    },
    {
      "epoch": 0.15043394406943106,
      "grad_norm": 1.9812514781951904,
      "learning_rate": 1.7128072445019407e-05,
      "loss": 1.0449,
      "num_input_tokens_seen": 13942400,
      "step": 117,
      "train_runtime": 1331.2362,
      "train_tokens_per_second": 10473.273
    },
    {
      "epoch": 0.1517197042751527,
      "grad_norm": 1.8844302892684937,
      "learning_rate": 1.7102199223803365e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 14069888,
      "step": 118,
      "train_runtime": 1342.8933,
      "train_tokens_per_second": 10477.294
    },
    {
      "epoch": 0.15300546448087432,
      "grad_norm": 1.816438913345337,
      "learning_rate": 1.7076326002587325e-05,
      "loss": 1.1185,
      "num_input_tokens_seen": 14175328,
      "step": 119,
      "train_runtime": 1352.3432,
      "train_tokens_per_second": 10482.049
    },
    {
      "epoch": 0.15429122468659595,
      "grad_norm": 1.8164008855819702,
      "learning_rate": 1.7050452781371283e-05,
      "loss": 1.1141,
      "num_input_tokens_seen": 14316736,
      "step": 120,
      "train_runtime": 1365.1583,
      "train_tokens_per_second": 10487.235
    },
    {
      "epoch": 0.15557698489231758,
      "grad_norm": 1.9856698513031006,
      "learning_rate": 1.702457956015524e-05,
      "loss": 1.0548,
      "num_input_tokens_seen": 14398720,
      "step": 121,
      "train_runtime": 1388.1489,
      "train_tokens_per_second": 10372.605
    },
    {
      "epoch": 0.1568627450980392,
      "grad_norm": 1.7079716920852661,
      "learning_rate": 1.69987063389392e-05,
      "loss": 1.1154,
      "num_input_tokens_seen": 14510944,
      "step": 122,
      "train_runtime": 1398.0002,
      "train_tokens_per_second": 10379.787
    },
    {
      "epoch": 0.15814850530376084,
      "grad_norm": 2.0213663578033447,
      "learning_rate": 1.6972833117723158e-05,
      "loss": 1.1813,
      "num_input_tokens_seen": 14627488,
      "step": 123,
      "train_runtime": 1408.2243,
      "train_tokens_per_second": 10387.186
    },
    {
      "epoch": 0.15943426550948248,
      "grad_norm": 1.5762097835540771,
      "learning_rate": 1.6946959896507115e-05,
      "loss": 0.8798,
      "num_input_tokens_seen": 14765056,
      "step": 124,
      "train_runtime": 1420.9688,
      "train_tokens_per_second": 10390.838
    },
    {
      "epoch": 0.1607200257152041,
      "grad_norm": 1.9591020345687866,
      "learning_rate": 1.6921086675291072e-05,
      "loss": 1.0846,
      "num_input_tokens_seen": 14870688,
      "step": 125,
      "train_runtime": 1430.1584,
      "train_tokens_per_second": 10397.931
    },
    {
      "epoch": 0.16200578592092574,
      "grad_norm": 1.9856301546096802,
      "learning_rate": 1.6895213454075033e-05,
      "loss": 1.1124,
      "num_input_tokens_seen": 15003424,
      "step": 126,
      "train_runtime": 1442.2654,
      "train_tokens_per_second": 10402.679
    },
    {
      "epoch": 0.16329154612664737,
      "grad_norm": 1.8877084255218506,
      "learning_rate": 1.6869340232858994e-05,
      "loss": 1.102,
      "num_input_tokens_seen": 15167424,
      "step": 127,
      "train_runtime": 1457.8734,
      "train_tokens_per_second": 10403.801
    },
    {
      "epoch": 0.164577306332369,
      "grad_norm": 1.8509068489074707,
      "learning_rate": 1.684346701164295e-05,
      "loss": 1.0739,
      "num_input_tokens_seen": 15348224,
      "step": 128,
      "train_runtime": 1475.0417,
      "train_tokens_per_second": 10405.282
    },
    {
      "epoch": 0.16586306653809066,
      "grad_norm": 1.9316065311431885,
      "learning_rate": 1.6817593790426908e-05,
      "loss": 1.1319,
      "num_input_tokens_seen": 15459680,
      "step": 129,
      "train_runtime": 1484.5083,
      "train_tokens_per_second": 10414.007
    },
    {
      "epoch": 0.1671488267438123,
      "grad_norm": 1.8838239908218384,
      "learning_rate": 1.679172056921087e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 15599488,
      "step": 130,
      "train_runtime": 1497.2084,
      "train_tokens_per_second": 10419.049
    },
    {
      "epoch": 0.16843458694953392,
      "grad_norm": 1.910106897354126,
      "learning_rate": 1.6765847347994826e-05,
      "loss": 1.1367,
      "num_input_tokens_seen": 15693728,
      "step": 131,
      "train_runtime": 1505.1617,
      "train_tokens_per_second": 10426.606
    },
    {
      "epoch": 0.16972034715525555,
      "grad_norm": 1.8794375658035278,
      "learning_rate": 1.6739974126778787e-05,
      "loss": 1.0579,
      "num_input_tokens_seen": 15827776,
      "step": 132,
      "train_runtime": 1517.1331,
      "train_tokens_per_second": 10432.688
    },
    {
      "epoch": 0.17100610736097718,
      "grad_norm": 1.9578567743301392,
      "learning_rate": 1.6714100905562744e-05,
      "loss": 0.9833,
      "num_input_tokens_seen": 15915456,
      "step": 133,
      "train_runtime": 1524.5049,
      "train_tokens_per_second": 10439.754
    },
    {
      "epoch": 0.17229186756669881,
      "grad_norm": 1.8861867189407349,
      "learning_rate": 1.66882276843467e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 16006112,
      "step": 134,
      "train_runtime": 1532.0904,
      "train_tokens_per_second": 10447.237
    },
    {
      "epoch": 0.17357762777242045,
      "grad_norm": 1.8860288858413696,
      "learning_rate": 1.6662354463130662e-05,
      "loss": 1.1243,
      "num_input_tokens_seen": 16105472,
      "step": 135,
      "train_runtime": 1540.4944,
      "train_tokens_per_second": 10454.742
    },
    {
      "epoch": 0.17486338797814208,
      "grad_norm": 1.7597628831863403,
      "learning_rate": 1.663648124191462e-05,
      "loss": 1.065,
      "num_input_tokens_seen": 16227296,
      "step": 136,
      "train_runtime": 1551.1796,
      "train_tokens_per_second": 10461.262
    },
    {
      "epoch": 0.1761491481838637,
      "grad_norm": 1.8316805362701416,
      "learning_rate": 1.6610608020698577e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 16327072,
      "step": 137,
      "train_runtime": 1559.8785,
      "train_tokens_per_second": 10466.887
    },
    {
      "epoch": 0.17743490838958534,
      "grad_norm": 1.965746283531189,
      "learning_rate": 1.6584734799482537e-05,
      "loss": 1.1353,
      "num_input_tokens_seen": 16424576,
      "step": 138,
      "train_runtime": 1568.1636,
      "train_tokens_per_second": 10473.764
    },
    {
      "epoch": 0.17872066859530697,
      "grad_norm": 1.7702914476394653,
      "learning_rate": 1.6558861578266498e-05,
      "loss": 1.1117,
      "num_input_tokens_seen": 16573120,
      "step": 139,
      "train_runtime": 1581.7542,
      "train_tokens_per_second": 10477.684
    },
    {
      "epoch": 0.1800064288010286,
      "grad_norm": 1.8150639533996582,
      "learning_rate": 1.6532988357050455e-05,
      "loss": 1.129,
      "num_input_tokens_seen": 16669440,
      "step": 140,
      "train_runtime": 1589.8087,
      "train_tokens_per_second": 10485.186
    },
    {
      "epoch": 0.18129218900675023,
      "grad_norm": 1.8353360891342163,
      "learning_rate": 1.6507115135834412e-05,
      "loss": 1.1148,
      "num_input_tokens_seen": 16771040,
      "step": 141,
      "train_runtime": 1598.5428,
      "train_tokens_per_second": 10491.455
    },
    {
      "epoch": 0.18257794921247186,
      "grad_norm": 2.064567804336548,
      "learning_rate": 1.648124191461837e-05,
      "loss": 1.0626,
      "num_input_tokens_seen": 16902048,
      "step": 142,
      "train_runtime": 1609.9524,
      "train_tokens_per_second": 10498.477
    },
    {
      "epoch": 0.1838637094181935,
      "grad_norm": 1.8049488067626953,
      "learning_rate": 1.645536869340233e-05,
      "loss": 1.1076,
      "num_input_tokens_seen": 17043808,
      "step": 143,
      "train_runtime": 1622.5917,
      "train_tokens_per_second": 10504.064
    },
    {
      "epoch": 0.18514946962391515,
      "grad_norm": 1.751596212387085,
      "learning_rate": 1.6429495472186288e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 17140096,
      "step": 144,
      "train_runtime": 1630.7355,
      "train_tokens_per_second": 10510.653
    },
    {
      "epoch": 0.18643522982963678,
      "grad_norm": 1.855686068534851,
      "learning_rate": 1.6403622250970248e-05,
      "loss": 1.1758,
      "num_input_tokens_seen": 17238624,
      "step": 145,
      "train_runtime": 1639.2446,
      "train_tokens_per_second": 10516.2
    },
    {
      "epoch": 0.18772099003535841,
      "grad_norm": 1.9099684953689575,
      "learning_rate": 1.6377749029754206e-05,
      "loss": 1.0719,
      "num_input_tokens_seen": 17360128,
      "step": 146,
      "train_runtime": 1650.2244,
      "train_tokens_per_second": 10519.859
    },
    {
      "epoch": 0.18900675024108005,
      "grad_norm": 1.8072643280029297,
      "learning_rate": 1.6351875808538166e-05,
      "loss": 1.1247,
      "num_input_tokens_seen": 17494240,
      "step": 147,
      "train_runtime": 1662.4473,
      "train_tokens_per_second": 10523.185
    },
    {
      "epoch": 0.19029251044680168,
      "grad_norm": 1.864859938621521,
      "learning_rate": 1.6326002587322123e-05,
      "loss": 1.0562,
      "num_input_tokens_seen": 17584160,
      "step": 148,
      "train_runtime": 1670.1325,
      "train_tokens_per_second": 10528.602
    },
    {
      "epoch": 0.1915782706525233,
      "grad_norm": 1.8772127628326416,
      "learning_rate": 1.630012936610608e-05,
      "loss": 1.0535,
      "num_input_tokens_seen": 17670144,
      "step": 149,
      "train_runtime": 1677.3621,
      "train_tokens_per_second": 10534.484
    },
    {
      "epoch": 0.19286403085824494,
      "grad_norm": 1.7363184690475464,
      "learning_rate": 1.6274256144890038e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 17771456,
      "step": 150,
      "train_runtime": 1686.1324,
      "train_tokens_per_second": 10539.775
    },
    {
      "epoch": 0.19414979106396657,
      "grad_norm": 1.9292726516723633,
      "learning_rate": 1.6248382923674e-05,
      "loss": 1.186,
      "num_input_tokens_seen": 17875200,
      "step": 151,
      "train_runtime": 1711.1828,
      "train_tokens_per_second": 10446.108
    },
    {
      "epoch": 0.1954355512696882,
      "grad_norm": 1.8201920986175537,
      "learning_rate": 1.622250970245796e-05,
      "loss": 1.0198,
      "num_input_tokens_seen": 17978048,
      "step": 152,
      "train_runtime": 1719.8774,
      "train_tokens_per_second": 10453.098
    },
    {
      "epoch": 0.19672131147540983,
      "grad_norm": 1.9103111028671265,
      "learning_rate": 1.6196636481241917e-05,
      "loss": 1.0822,
      "num_input_tokens_seen": 18092224,
      "step": 153,
      "train_runtime": 1730.076,
      "train_tokens_per_second": 10457.473
    },
    {
      "epoch": 0.19800707168113146,
      "grad_norm": 1.990533709526062,
      "learning_rate": 1.6170763260025874e-05,
      "loss": 1.1262,
      "num_input_tokens_seen": 18180064,
      "step": 154,
      "train_runtime": 1737.4747,
      "train_tokens_per_second": 10463.499
    },
    {
      "epoch": 0.1992928318868531,
      "grad_norm": 1.7888742685317993,
      "learning_rate": 1.6144890038809835e-05,
      "loss": 1.1955,
      "num_input_tokens_seen": 18265056,
      "step": 155,
      "train_runtime": 1744.6485,
      "train_tokens_per_second": 10469.19
    },
    {
      "epoch": 0.20057859209257473,
      "grad_norm": 1.7337497472763062,
      "learning_rate": 1.6119016817593792e-05,
      "loss": 1.0541,
      "num_input_tokens_seen": 18367328,
      "step": 156,
      "train_runtime": 1753.372,
      "train_tokens_per_second": 10475.431
    },
    {
      "epoch": 0.20057859209257473,
      "eval_loss": 1.1330561637878418,
      "eval_runtime": 22.3584,
      "eval_samples_per_second": 44.458,
      "eval_steps_per_second": 1.431,
      "num_input_tokens_seen": 18367328,
      "step": 156
    },
    {
      "epoch": 0.20186435229829636,
      "grad_norm": 1.760105848312378,
      "learning_rate": 1.609314359637775e-05,
      "loss": 1.092,
      "num_input_tokens_seen": 18469024,
      "step": 157,
      "train_runtime": 1784.5882,
      "train_tokens_per_second": 10349.179
    },
    {
      "epoch": 0.203150112504018,
      "grad_norm": 2.001173496246338,
      "learning_rate": 1.606727037516171e-05,
      "loss": 1.15,
      "num_input_tokens_seen": 18603648,
      "step": 158,
      "train_runtime": 1796.907,
      "train_tokens_per_second": 10353.15
    },
    {
      "epoch": 0.20443587270973965,
      "grad_norm": 1.806795358657837,
      "learning_rate": 1.6041397153945667e-05,
      "loss": 1.024,
      "num_input_tokens_seen": 18693536,
      "step": 159,
      "train_runtime": 1804.6539,
      "train_tokens_per_second": 10358.516
    },
    {
      "epoch": 0.20572163291546128,
      "grad_norm": 1.9059035778045654,
      "learning_rate": 1.6015523932729628e-05,
      "loss": 1.1534,
      "num_input_tokens_seen": 18854592,
      "step": 160,
      "train_runtime": 1819.9574,
      "train_tokens_per_second": 10359.908
    },
    {
      "epoch": 0.2070073931211829,
      "grad_norm": 1.902287483215332,
      "learning_rate": 1.5989650711513585e-05,
      "loss": 0.9808,
      "num_input_tokens_seen": 18972384,
      "step": 161,
      "train_runtime": 1830.4631,
      "train_tokens_per_second": 10364.8
    },
    {
      "epoch": 0.20829315332690454,
      "grad_norm": 1.9465879201889038,
      "learning_rate": 1.5963777490297542e-05,
      "loss": 1.1595,
      "num_input_tokens_seen": 19119424,
      "step": 162,
      "train_runtime": 1844.4228,
      "train_tokens_per_second": 10366.075
    },
    {
      "epoch": 0.20957891353262617,
      "grad_norm": 1.8765859603881836,
      "learning_rate": 1.59379042690815e-05,
      "loss": 1.1413,
      "num_input_tokens_seen": 19229984,
      "step": 163,
      "train_runtime": 1854.1827,
      "train_tokens_per_second": 10371.138
    },
    {
      "epoch": 0.2108646737383478,
      "grad_norm": 1.9399867057800293,
      "learning_rate": 1.591203104786546e-05,
      "loss": 1.0859,
      "num_input_tokens_seen": 19334016,
      "step": 164,
      "train_runtime": 1863.2733,
      "train_tokens_per_second": 10376.372
    },
    {
      "epoch": 0.21215043394406943,
      "grad_norm": 1.999969244003296,
      "learning_rate": 1.588615782664942e-05,
      "loss": 1.1394,
      "num_input_tokens_seen": 19501312,
      "step": 165,
      "train_runtime": 1879.0538,
      "train_tokens_per_second": 10378.261
    },
    {
      "epoch": 0.21343619414979106,
      "grad_norm": 1.8910810947418213,
      "learning_rate": 1.5860284605433378e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 19602976,
      "step": 166,
      "train_runtime": 1887.7727,
      "train_tokens_per_second": 10384.182
    },
    {
      "epoch": 0.2147219543555127,
      "grad_norm": 1.824230432510376,
      "learning_rate": 1.5834411384217335e-05,
      "loss": 1.0689,
      "num_input_tokens_seen": 19727904,
      "step": 167,
      "train_runtime": 1898.9411,
      "train_tokens_per_second": 10388.897
    },
    {
      "epoch": 0.21600771456123433,
      "grad_norm": 1.9098621606826782,
      "learning_rate": 1.5808538163001296e-05,
      "loss": 1.069,
      "num_input_tokens_seen": 19844096,
      "step": 168,
      "train_runtime": 1909.7001,
      "train_tokens_per_second": 10391.211
    },
    {
      "epoch": 0.21729347476695596,
      "grad_norm": 1.773947834968567,
      "learning_rate": 1.5782664941785253e-05,
      "loss": 1.0426,
      "num_input_tokens_seen": 19969248,
      "step": 169,
      "train_runtime": 1920.8606,
      "train_tokens_per_second": 10395.99
    },
    {
      "epoch": 0.2185792349726776,
      "grad_norm": 1.7338457107543945,
      "learning_rate": 1.575679172056921e-05,
      "loss": 1.0428,
      "num_input_tokens_seen": 20165376,
      "step": 170,
      "train_runtime": 1940.1892,
      "train_tokens_per_second": 10393.51
    },
    {
      "epoch": 0.21986499517839922,
      "grad_norm": 1.7944598197937012,
      "learning_rate": 1.573091849935317e-05,
      "loss": 1.1008,
      "num_input_tokens_seen": 20282464,
      "step": 171,
      "train_runtime": 1950.6375,
      "train_tokens_per_second": 10397.864
    },
    {
      "epoch": 0.22115075538412085,
      "grad_norm": 1.7881031036376953,
      "learning_rate": 1.570504527813713e-05,
      "loss": 1.0998,
      "num_input_tokens_seen": 20414304,
      "step": 172,
      "train_runtime": 1962.9682,
      "train_tokens_per_second": 10399.712
    },
    {
      "epoch": 0.22243651558984248,
      "grad_norm": 1.8311177492141724,
      "learning_rate": 1.567917205692109e-05,
      "loss": 1.1359,
      "num_input_tokens_seen": 20524000,
      "step": 173,
      "train_runtime": 1972.6273,
      "train_tokens_per_second": 10404.398
    },
    {
      "epoch": 0.22372227579556414,
      "grad_norm": 1.774649739265442,
      "learning_rate": 1.5653298835705046e-05,
      "loss": 1.1199,
      "num_input_tokens_seen": 20615872,
      "step": 174,
      "train_runtime": 1980.5452,
      "train_tokens_per_second": 10409.191
    },
    {
      "epoch": 0.22500803600128577,
      "grad_norm": 1.8202511072158813,
      "learning_rate": 1.5627425614489004e-05,
      "loss": 1.1552,
      "num_input_tokens_seen": 20737184,
      "step": 175,
      "train_runtime": 1991.4125,
      "train_tokens_per_second": 10413.304
    },
    {
      "epoch": 0.2262937962070074,
      "grad_norm": 1.8184281587600708,
      "learning_rate": 1.5601552393272964e-05,
      "loss": 1.0945,
      "num_input_tokens_seen": 20871712,
      "step": 176,
      "train_runtime": 2003.8054,
      "train_tokens_per_second": 10416.038
    },
    {
      "epoch": 0.22757955641272903,
      "grad_norm": 2.0196473598480225,
      "learning_rate": 1.5575679172056925e-05,
      "loss": 1.0248,
      "num_input_tokens_seen": 21029760,
      "step": 177,
      "train_runtime": 2018.6807,
      "train_tokens_per_second": 10417.576
    },
    {
      "epoch": 0.22886531661845066,
      "grad_norm": 1.8468624353408813,
      "learning_rate": 1.5549805950840882e-05,
      "loss": 1.1439,
      "num_input_tokens_seen": 21118112,
      "step": 178,
      "train_runtime": 2026.2014,
      "train_tokens_per_second": 10422.514
    },
    {
      "epoch": 0.2301510768241723,
      "grad_norm": 2.0327556133270264,
      "learning_rate": 1.552393272962484e-05,
      "loss": 1.0553,
      "num_input_tokens_seen": 21217408,
      "step": 179,
      "train_runtime": 2034.8324,
      "train_tokens_per_second": 10427.104
    },
    {
      "epoch": 0.23143683702989393,
      "grad_norm": 1.9502851963043213,
      "learning_rate": 1.5498059508408797e-05,
      "loss": 1.0866,
      "num_input_tokens_seen": 21328928,
      "step": 180,
      "train_runtime": 2044.8024,
      "train_tokens_per_second": 10430.802
    },
    {
      "epoch": 0.23272259723561556,
      "grad_norm": 1.8489340543746948,
      "learning_rate": 1.5472186287192757e-05,
      "loss": 1.1345,
      "num_input_tokens_seen": 21433248,
      "step": 181,
      "train_runtime": 2072.5697,
      "train_tokens_per_second": 10341.388
    },
    {
      "epoch": 0.2340083574413372,
      "grad_norm": 2.063244104385376,
      "learning_rate": 1.5446313065976715e-05,
      "loss": 1.0425,
      "num_input_tokens_seen": 21543552,
      "step": 182,
      "train_runtime": 2082.003,
      "train_tokens_per_second": 10347.513
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 1.8250895738601685,
      "learning_rate": 1.5420439844760672e-05,
      "loss": 1.1022,
      "num_input_tokens_seen": 21631584,
      "step": 183,
      "train_runtime": 2089.5086,
      "train_tokens_per_second": 10352.474
    },
    {
      "epoch": 0.23657987785278045,
      "grad_norm": 1.8058557510375977,
      "learning_rate": 1.5394566623544633e-05,
      "loss": 1.1283,
      "num_input_tokens_seen": 21732768,
      "step": 184,
      "train_runtime": 2098.3876,
      "train_tokens_per_second": 10356.889
    },
    {
      "epoch": 0.23786563805850208,
      "grad_norm": 1.8823782205581665,
      "learning_rate": 1.5368693402328593e-05,
      "loss": 1.0454,
      "num_input_tokens_seen": 21847392,
      "step": 185,
      "train_runtime": 2108.5281,
      "train_tokens_per_second": 10361.442
    },
    {
      "epoch": 0.2391513982642237,
      "grad_norm": 1.9651679992675781,
      "learning_rate": 1.534282018111255e-05,
      "loss": 1.1543,
      "num_input_tokens_seen": 21969408,
      "step": 186,
      "train_runtime": 2119.7121,
      "train_tokens_per_second": 10364.336
    },
    {
      "epoch": 0.24043715846994534,
      "grad_norm": 1.9040675163269043,
      "learning_rate": 1.5316946959896508e-05,
      "loss": 1.1195,
      "num_input_tokens_seen": 22082176,
      "step": 187,
      "train_runtime": 2129.7977,
      "train_tokens_per_second": 10368.204
    },
    {
      "epoch": 0.24172291867566698,
      "grad_norm": 1.8690330982208252,
      "learning_rate": 1.5291073738680465e-05,
      "loss": 1.1459,
      "num_input_tokens_seen": 22172480,
      "step": 188,
      "train_runtime": 2137.5122,
      "train_tokens_per_second": 10373.031
    },
    {
      "epoch": 0.24300867888138863,
      "grad_norm": 1.873816967010498,
      "learning_rate": 1.5265200517464426e-05,
      "loss": 1.1256,
      "num_input_tokens_seen": 22305728,
      "step": 189,
      "train_runtime": 2149.436,
      "train_tokens_per_second": 10377.479
    },
    {
      "epoch": 0.24429443908711027,
      "grad_norm": 1.936040997505188,
      "learning_rate": 1.5239327296248385e-05,
      "loss": 1.0522,
      "num_input_tokens_seen": 22460672,
      "step": 190,
      "train_runtime": 2163.9753,
      "train_tokens_per_second": 10379.357
    },
    {
      "epoch": 0.2455801992928319,
      "grad_norm": 1.8445260524749756,
      "learning_rate": 1.5213454075032344e-05,
      "loss": 1.1412,
      "num_input_tokens_seen": 22556352,
      "step": 191,
      "train_runtime": 2172.3429,
      "train_tokens_per_second": 10383.421
    },
    {
      "epoch": 0.24686595949855353,
      "grad_norm": 1.9163671731948853,
      "learning_rate": 1.5187580853816301e-05,
      "loss": 1.1672,
      "num_input_tokens_seen": 22662624,
      "step": 192,
      "train_runtime": 2181.6935,
      "train_tokens_per_second": 10387.63
    },
    {
      "epoch": 0.24815171970427516,
      "grad_norm": 2.0177226066589355,
      "learning_rate": 1.516170763260026e-05,
      "loss": 1.1193,
      "num_input_tokens_seen": 22789664,
      "step": 193,
      "train_runtime": 2193.5134,
      "train_tokens_per_second": 10389.571
    },
    {
      "epoch": 0.2494374799099968,
      "grad_norm": 1.8356866836547852,
      "learning_rate": 1.5135834411384219e-05,
      "loss": 1.1705,
      "num_input_tokens_seen": 22878816,
      "step": 194,
      "train_runtime": 2201.1512,
      "train_tokens_per_second": 10394.023
    },
    {
      "epoch": 0.2507232401157184,
      "grad_norm": 1.9464590549468994,
      "learning_rate": 1.5109961190168176e-05,
      "loss": 1.2413,
      "num_input_tokens_seen": 23036256,
      "step": 195,
      "train_runtime": 2216.0262,
      "train_tokens_per_second": 10395.299
    },
    {
      "epoch": 0.25200900032144,
      "grad_norm": 1.9846364259719849,
      "learning_rate": 1.5084087968952135e-05,
      "loss": 1.1349,
      "num_input_tokens_seen": 23113184,
      "step": 196,
      "train_runtime": 2222.6223,
      "train_tokens_per_second": 10399.061
    },
    {
      "epoch": 0.2532947605271617,
      "grad_norm": 1.921323537826538,
      "learning_rate": 1.5058214747736096e-05,
      "loss": 1.0196,
      "num_input_tokens_seen": 23227616,
      "step": 197,
      "train_runtime": 2232.9722,
      "train_tokens_per_second": 10402.107
    },
    {
      "epoch": 0.25458052073288334,
      "grad_norm": 1.7825725078582764,
      "learning_rate": 1.5032341526520053e-05,
      "loss": 1.0327,
      "num_input_tokens_seen": 23394208,
      "step": 198,
      "train_runtime": 2249.1819,
      "train_tokens_per_second": 10401.208
    },
    {
      "epoch": 0.255866280938605,
      "grad_norm": 2.1068551540374756,
      "learning_rate": 1.5006468305304012e-05,
      "loss": 1.0875,
      "num_input_tokens_seen": 23528416,
      "step": 199,
      "train_runtime": 2261.1565,
      "train_tokens_per_second": 10405.479
    },
    {
      "epoch": 0.2571520411443266,
      "grad_norm": 1.8937175273895264,
      "learning_rate": 1.498059508408797e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 23670240,
      "step": 200,
      "train_runtime": 2274.3795,
      "train_tokens_per_second": 10407.34
    },
    {
      "epoch": 0.25843780135004824,
      "grad_norm": 1.8117809295654297,
      "learning_rate": 1.4954721862871928e-05,
      "loss": 0.98,
      "num_input_tokens_seen": 23795712,
      "step": 201,
      "train_runtime": 2285.6253,
      "train_tokens_per_second": 10411.029
    },
    {
      "epoch": 0.25972356155576987,
      "grad_norm": 1.9877797365188599,
      "learning_rate": 1.4928848641655887e-05,
      "loss": 1.1332,
      "num_input_tokens_seen": 23913248,
      "step": 202,
      "train_runtime": 2295.9518,
      "train_tokens_per_second": 10415.396
    },
    {
      "epoch": 0.2610093217614915,
      "grad_norm": 1.832665205001831,
      "learning_rate": 1.4902975420439846e-05,
      "loss": 1.0939,
      "num_input_tokens_seen": 24017152,
      "step": 203,
      "train_runtime": 2305.1109,
      "train_tokens_per_second": 10419.088
    },
    {
      "epoch": 0.26229508196721313,
      "grad_norm": 1.8349584341049194,
      "learning_rate": 1.4877102199223805e-05,
      "loss": 1.1451,
      "num_input_tokens_seen": 24132384,
      "step": 204,
      "train_runtime": 2315.2627,
      "train_tokens_per_second": 10423.173
    },
    {
      "epoch": 0.26358084217293476,
      "grad_norm": 1.8495604991912842,
      "learning_rate": 1.4851228978007764e-05,
      "loss": 1.0344,
      "num_input_tokens_seen": 24217024,
      "step": 205,
      "train_runtime": 2322.5078,
      "train_tokens_per_second": 10427.101
    },
    {
      "epoch": 0.2648666023786564,
      "grad_norm": 1.7923839092254639,
      "learning_rate": 1.4825355756791721e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 24329984,
      "step": 206,
      "train_runtime": 2332.827,
      "train_tokens_per_second": 10429.399
    },
    {
      "epoch": 0.266152362584378,
      "grad_norm": 1.8214589357376099,
      "learning_rate": 1.479948253557568e-05,
      "loss": 1.091,
      "num_input_tokens_seen": 24451776,
      "step": 207,
      "train_runtime": 2343.5108,
      "train_tokens_per_second": 10433.823
    },
    {
      "epoch": 0.26743812279009965,
      "grad_norm": 1.9028258323669434,
      "learning_rate": 1.4773609314359637e-05,
      "loss": 1.1523,
      "num_input_tokens_seen": 24630016,
      "step": 208,
      "train_runtime": 2360.69,
      "train_tokens_per_second": 10433.397
    },
    {
      "epoch": 0.2687238829958213,
      "grad_norm": 1.80238676071167,
      "learning_rate": 1.4747736093143598e-05,
      "loss": 1.0408,
      "num_input_tokens_seen": 24752800,
      "step": 209,
      "train_runtime": 2371.5796,
      "train_tokens_per_second": 10437.263
    },
    {
      "epoch": 0.2700096432015429,
      "grad_norm": 1.8515748977661133,
      "learning_rate": 1.4721862871927557e-05,
      "loss": 1.0605,
      "num_input_tokens_seen": 24844864,
      "step": 210,
      "train_runtime": 2379.4213,
      "train_tokens_per_second": 10441.557
    },
    {
      "epoch": 0.27129540340726455,
      "grad_norm": 1.831057071685791,
      "learning_rate": 1.4695989650711514e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 25002304,
      "step": 211,
      "train_runtime": 2412.4948,
      "train_tokens_per_second": 10363.672
    },
    {
      "epoch": 0.2725811636129862,
      "grad_norm": 2.014904260635376,
      "learning_rate": 1.4670116429495473e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 25133536,
      "step": 212,
      "train_runtime": 2424.3186,
      "train_tokens_per_second": 10367.258
    },
    {
      "epoch": 0.2738669238187078,
      "grad_norm": 1.9899771213531494,
      "learning_rate": 1.4644243208279432e-05,
      "loss": 1.0888,
      "num_input_tokens_seen": 25302464,
      "step": 213,
      "train_runtime": 2440.569,
      "train_tokens_per_second": 10367.445
    },
    {
      "epoch": 0.27515268402442944,
      "grad_norm": 1.9485491514205933,
      "learning_rate": 1.461836998706339e-05,
      "loss": 1.0257,
      "num_input_tokens_seen": 25400640,
      "step": 214,
      "train_runtime": 2449.2043,
      "train_tokens_per_second": 10370.976
    },
    {
      "epoch": 0.27643844423015107,
      "grad_norm": 1.7471483945846558,
      "learning_rate": 1.4592496765847349e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 25491296,
      "step": 215,
      "train_runtime": 2456.9876,
      "train_tokens_per_second": 10375.02
    },
    {
      "epoch": 0.2777242044358727,
      "grad_norm": 1.8240132331848145,
      "learning_rate": 1.456662354463131e-05,
      "loss": 1.0384,
      "num_input_tokens_seen": 25584320,
      "step": 216,
      "train_runtime": 2464.9704,
      "train_tokens_per_second": 10379.159
    },
    {
      "epoch": 0.27900996464159433,
      "grad_norm": 1.8008581399917603,
      "learning_rate": 1.4540750323415266e-05,
      "loss": 1.0473,
      "num_input_tokens_seen": 25675328,
      "step": 217,
      "train_runtime": 2472.784,
      "train_tokens_per_second": 10383.166
    },
    {
      "epoch": 0.28029572484731596,
      "grad_norm": 1.8560556173324585,
      "learning_rate": 1.4514877102199225e-05,
      "loss": 1.0333,
      "num_input_tokens_seen": 25761216,
      "step": 218,
      "train_runtime": 2480.13,
      "train_tokens_per_second": 10387.043
    },
    {
      "epoch": 0.2815814850530376,
      "grad_norm": 1.9014114141464233,
      "learning_rate": 1.4489003880983183e-05,
      "loss": 1.1028,
      "num_input_tokens_seen": 25890592,
      "step": 219,
      "train_runtime": 2491.8423,
      "train_tokens_per_second": 10390.141
    },
    {
      "epoch": 0.2828672452587592,
      "grad_norm": 1.8335984945297241,
      "learning_rate": 1.4463130659767142e-05,
      "loss": 1.045,
      "num_input_tokens_seen": 25970880,
      "step": 220,
      "train_runtime": 2498.6975,
      "train_tokens_per_second": 10393.767
    },
    {
      "epoch": 0.28415300546448086,
      "grad_norm": 1.9171757698059082,
      "learning_rate": 1.44372574385511e-05,
      "loss": 1.074,
      "num_input_tokens_seen": 26051808,
      "step": 221,
      "train_runtime": 2505.5849,
      "train_tokens_per_second": 10397.496
    },
    {
      "epoch": 0.2854387656702025,
      "grad_norm": 1.849561333656311,
      "learning_rate": 1.4411384217335061e-05,
      "loss": 1.0662,
      "num_input_tokens_seen": 26163232,
      "step": 222,
      "train_runtime": 2515.3859,
      "train_tokens_per_second": 10401.28
    },
    {
      "epoch": 0.2867245258759241,
      "grad_norm": 1.886615514755249,
      "learning_rate": 1.4385510996119019e-05,
      "loss": 1.1053,
      "num_input_tokens_seen": 26257664,
      "step": 223,
      "train_runtime": 2523.536,
      "train_tokens_per_second": 10405.108
    },
    {
      "epoch": 0.28801028608164575,
      "grad_norm": 1.7920629978179932,
      "learning_rate": 1.4359637774902977e-05,
      "loss": 1.0041,
      "num_input_tokens_seen": 26368288,
      "step": 224,
      "train_runtime": 2533.3464,
      "train_tokens_per_second": 10408.481
    },
    {
      "epoch": 0.2892960462873674,
      "grad_norm": 1.9786632061004639,
      "learning_rate": 1.4333764553686935e-05,
      "loss": 1.1993,
      "num_input_tokens_seen": 26482400,
      "step": 225,
      "train_runtime": 2543.3441,
      "train_tokens_per_second": 10412.433
    },
    {
      "epoch": 0.290581806493089,
      "grad_norm": 1.9302722215652466,
      "learning_rate": 1.4307891332470894e-05,
      "loss": 1.0486,
      "num_input_tokens_seen": 26580512,
      "step": 226,
      "train_runtime": 2551.839,
      "train_tokens_per_second": 10416.218
    },
    {
      "epoch": 0.2918675666988107,
      "grad_norm": 1.7275605201721191,
      "learning_rate": 1.4282018111254851e-05,
      "loss": 0.9721,
      "num_input_tokens_seen": 26699392,
      "step": 227,
      "train_runtime": 2562.4385,
      "train_tokens_per_second": 10419.525
    },
    {
      "epoch": 0.29315332690453233,
      "grad_norm": 1.8996021747589111,
      "learning_rate": 1.425614489003881e-05,
      "loss": 1.0883,
      "num_input_tokens_seen": 26856992,
      "step": 228,
      "train_runtime": 2577.1118,
      "train_tokens_per_second": 10421.353
    },
    {
      "epoch": 0.29443908711025396,
      "grad_norm": 1.8853155374526978,
      "learning_rate": 1.423027166882277e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 26964960,
      "step": 229,
      "train_runtime": 2586.5183,
      "train_tokens_per_second": 10425.196
    },
    {
      "epoch": 0.2957248473159756,
      "grad_norm": 1.9568660259246826,
      "learning_rate": 1.4204398447606728e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 27099008,
      "step": 230,
      "train_runtime": 2599.0806,
      "train_tokens_per_second": 10426.382
    },
    {
      "epoch": 0.2970106075216972,
      "grad_norm": 1.8366093635559082,
      "learning_rate": 1.4178525226390687e-05,
      "loss": 1.0547,
      "num_input_tokens_seen": 27213760,
      "step": 231,
      "train_runtime": 2609.5848,
      "train_tokens_per_second": 10428.387
    },
    {
      "epoch": 0.29829636772741885,
      "grad_norm": 1.8327536582946777,
      "learning_rate": 1.4152652005174646e-05,
      "loss": 1.0903,
      "num_input_tokens_seen": 27297344,
      "step": 232,
      "train_runtime": 2616.7236,
      "train_tokens_per_second": 10431.879
    },
    {
      "epoch": 0.2995821279331405,
      "grad_norm": 1.8904881477355957,
      "learning_rate": 1.4126778783958603e-05,
      "loss": 1.0861,
      "num_input_tokens_seen": 27426688,
      "step": 233,
      "train_runtime": 2628.7609,
      "train_tokens_per_second": 10433.314
    },
    {
      "epoch": 0.3008678881388621,
      "grad_norm": 1.99700927734375,
      "learning_rate": 1.4100905562742562e-05,
      "loss": 1.0771,
      "num_input_tokens_seen": 27563648,
      "step": 234,
      "train_runtime": 2641.4168,
      "train_tokens_per_second": 10435.176
    },
    {
      "epoch": 0.3008678881388621,
      "eval_loss": 1.1202009916305542,
      "eval_runtime": 22.3846,
      "eval_samples_per_second": 44.406,
      "eval_steps_per_second": 1.43,
      "num_input_tokens_seen": 27563648,
      "step": 234
    },
    {
      "epoch": 0.30215364834458375,
      "grad_norm": 1.886243462562561,
      "learning_rate": 1.4075032341526523e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 27753376,
      "step": 235,
      "train_runtime": 2682.2789,
      "train_tokens_per_second": 10346.939
    },
    {
      "epoch": 0.3034394085503054,
      "grad_norm": 1.8130700588226318,
      "learning_rate": 1.404915912031048e-05,
      "loss": 1.0335,
      "num_input_tokens_seen": 27868672,
      "step": 236,
      "train_runtime": 2692.3246,
      "train_tokens_per_second": 10351.156
    },
    {
      "epoch": 0.304725168756027,
      "grad_norm": 1.870937466621399,
      "learning_rate": 1.4023285899094439e-05,
      "loss": 1.036,
      "num_input_tokens_seen": 27970848,
      "step": 237,
      "train_runtime": 2701.1905,
      "train_tokens_per_second": 10355.008
    },
    {
      "epoch": 0.30601092896174864,
      "grad_norm": 1.9421213865280151,
      "learning_rate": 1.3997412677878396e-05,
      "loss": 1.0861,
      "num_input_tokens_seen": 28103104,
      "step": 238,
      "train_runtime": 2713.4976,
      "train_tokens_per_second": 10356.783
    },
    {
      "epoch": 0.30729668916747027,
      "grad_norm": 2.0607948303222656,
      "learning_rate": 1.3971539456662355e-05,
      "loss": 1.1563,
      "num_input_tokens_seen": 28200704,
      "step": 239,
      "train_runtime": 2721.862,
      "train_tokens_per_second": 10360.813
    },
    {
      "epoch": 0.3085824493731919,
      "grad_norm": 1.9641258716583252,
      "learning_rate": 1.3945666235446314e-05,
      "loss": 1.0857,
      "num_input_tokens_seen": 28301088,
      "step": 240,
      "train_runtime": 2730.5823,
      "train_tokens_per_second": 10364.488
    },
    {
      "epoch": 0.30986820957891353,
      "grad_norm": 1.7765169143676758,
      "learning_rate": 1.3919793014230271e-05,
      "loss": 1.17,
      "num_input_tokens_seen": 28407104,
      "step": 241,
      "train_runtime": 2757.845,
      "train_tokens_per_second": 10300.472
    },
    {
      "epoch": 0.31115396978463516,
      "grad_norm": 1.8327834606170654,
      "learning_rate": 1.3893919793014232e-05,
      "loss": 1.1576,
      "num_input_tokens_seen": 28551616,
      "step": 242,
      "train_runtime": 2770.8246,
      "train_tokens_per_second": 10304.375
    },
    {
      "epoch": 0.3124397299903568,
      "grad_norm": 1.8959366083145142,
      "learning_rate": 1.3868046571798191e-05,
      "loss": 1.1029,
      "num_input_tokens_seen": 28676992,
      "step": 243,
      "train_runtime": 2781.7813,
      "train_tokens_per_second": 10308.859
    },
    {
      "epoch": 0.3137254901960784,
      "grad_norm": 1.906044840812683,
      "learning_rate": 1.3842173350582148e-05,
      "loss": 1.0936,
      "num_input_tokens_seen": 28808192,
      "step": 244,
      "train_runtime": 2793.4191,
      "train_tokens_per_second": 10312.879
    },
    {
      "epoch": 0.31501125040180006,
      "grad_norm": 1.6482162475585938,
      "learning_rate": 1.3816300129366107e-05,
      "loss": 1.0089,
      "num_input_tokens_seen": 28911680,
      "step": 245,
      "train_runtime": 2802.3761,
      "train_tokens_per_second": 10316.845
    },
    {
      "epoch": 0.3162970106075217,
      "grad_norm": 2.033006191253662,
      "learning_rate": 1.3790426908150065e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 29016864,
      "step": 246,
      "train_runtime": 2811.4614,
      "train_tokens_per_second": 10320.919
    },
    {
      "epoch": 0.3175827708132433,
      "grad_norm": 1.7948728799819946,
      "learning_rate": 1.3764553686934023e-05,
      "loss": 1.0618,
      "num_input_tokens_seen": 29131168,
      "step": 247,
      "train_runtime": 2821.6622,
      "train_tokens_per_second": 10324.116
    },
    {
      "epoch": 0.31886853101896495,
      "grad_norm": 1.90618097782135,
      "learning_rate": 1.3738680465717984e-05,
      "loss": 1.0781,
      "num_input_tokens_seen": 29220064,
      "step": 248,
      "train_runtime": 2829.145,
      "train_tokens_per_second": 10328.231
    },
    {
      "epoch": 0.3201542912246866,
      "grad_norm": 1.9356735944747925,
      "learning_rate": 1.3712807244501941e-05,
      "loss": 1.081,
      "num_input_tokens_seen": 29327360,
      "step": 249,
      "train_runtime": 2838.3079,
      "train_tokens_per_second": 10332.691
    },
    {
      "epoch": 0.3214400514304082,
      "grad_norm": 1.9027749300003052,
      "learning_rate": 1.36869340232859e-05,
      "loss": 1.0783,
      "num_input_tokens_seen": 29460416,
      "step": 250,
      "train_runtime": 2850.5438,
      "train_tokens_per_second": 10335.016
    },
    {
      "epoch": 0.32272581163612984,
      "grad_norm": 1.9512473344802856,
      "learning_rate": 1.366106080206986e-05,
      "loss": 1.1187,
      "num_input_tokens_seen": 29560960,
      "step": 251,
      "train_runtime": 2859.2337,
      "train_tokens_per_second": 10338.77
    },
    {
      "epoch": 0.3240115718418515,
      "grad_norm": 1.8211243152618408,
      "learning_rate": 1.3635187580853817e-05,
      "loss": 1.0958,
      "num_input_tokens_seen": 29663328,
      "step": 252,
      "train_runtime": 2868.0908,
      "train_tokens_per_second": 10342.535
    },
    {
      "epoch": 0.3252973320475731,
      "grad_norm": 1.9262733459472656,
      "learning_rate": 1.3609314359637776e-05,
      "loss": 1.0872,
      "num_input_tokens_seen": 29746208,
      "step": 253,
      "train_runtime": 2875.16,
      "train_tokens_per_second": 10345.931
    },
    {
      "epoch": 0.32658309225329474,
      "grad_norm": 1.8188444375991821,
      "learning_rate": 1.3583441138421733e-05,
      "loss": 1.0855,
      "num_input_tokens_seen": 29894432,
      "step": 254,
      "train_runtime": 2888.8665,
      "train_tokens_per_second": 10348.153
    },
    {
      "epoch": 0.32786885245901637,
      "grad_norm": 1.9136427640914917,
      "learning_rate": 1.3557567917205693e-05,
      "loss": 1.1348,
      "num_input_tokens_seen": 29994112,
      "step": 255,
      "train_runtime": 2897.5218,
      "train_tokens_per_second": 10351.643
    },
    {
      "epoch": 0.329154612664738,
      "grad_norm": 1.8818548917770386,
      "learning_rate": 1.3531694695989652e-05,
      "loss": 1.0323,
      "num_input_tokens_seen": 30130016,
      "step": 256,
      "train_runtime": 2910.0743,
      "train_tokens_per_second": 10353.693
    },
    {
      "epoch": 0.3304403728704597,
      "grad_norm": 1.8252167701721191,
      "learning_rate": 1.350582147477361e-05,
      "loss": 1.1093,
      "num_input_tokens_seen": 30265760,
      "step": 257,
      "train_runtime": 2922.5727,
      "train_tokens_per_second": 10355.862
    },
    {
      "epoch": 0.3317261330761813,
      "grad_norm": 1.7374950647354126,
      "learning_rate": 1.3479948253557569e-05,
      "loss": 1.0167,
      "num_input_tokens_seen": 30376608,
      "step": 258,
      "train_runtime": 2932.2957,
      "train_tokens_per_second": 10359.326
    },
    {
      "epoch": 0.33301189328190295,
      "grad_norm": 1.8809834718704224,
      "learning_rate": 1.3454075032341528e-05,
      "loss": 1.0811,
      "num_input_tokens_seen": 30516832,
      "step": 259,
      "train_runtime": 2945.3073,
      "train_tokens_per_second": 10361.171
    },
    {
      "epoch": 0.3342976534876246,
      "grad_norm": 1.8606772422790527,
      "learning_rate": 1.3428201811125485e-05,
      "loss": 1.1816,
      "num_input_tokens_seen": 30630464,
      "step": 260,
      "train_runtime": 2955.2838,
      "train_tokens_per_second": 10364.644
    },
    {
      "epoch": 0.3355834136933462,
      "grad_norm": 1.841683030128479,
      "learning_rate": 1.3402328589909446e-05,
      "loss": 1.0374,
      "num_input_tokens_seen": 30773344,
      "step": 261,
      "train_runtime": 2968.6865,
      "train_tokens_per_second": 10365.98
    },
    {
      "epoch": 0.33686917389906784,
      "grad_norm": 1.9608851671218872,
      "learning_rate": 1.3376455368693405e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 30899296,
      "step": 262,
      "train_runtime": 2980.4948,
      "train_tokens_per_second": 10367.17
    },
    {
      "epoch": 0.3381549341047895,
      "grad_norm": 1.9880949258804321,
      "learning_rate": 1.3350582147477362e-05,
      "loss": 1.0203,
      "num_input_tokens_seen": 31000256,
      "step": 263,
      "train_runtime": 2989.2311,
      "train_tokens_per_second": 10370.645
    },
    {
      "epoch": 0.3394406943105111,
      "grad_norm": 1.9502097368240356,
      "learning_rate": 1.332470892626132e-05,
      "loss": 1.1464,
      "num_input_tokens_seen": 31104768,
      "step": 264,
      "train_runtime": 2998.2398,
      "train_tokens_per_second": 10374.343
    },
    {
      "epoch": 0.34072645451623274,
      "grad_norm": 1.934061050415039,
      "learning_rate": 1.3298835705045278e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 31232832,
      "step": 265,
      "train_runtime": 3009.8265,
      "train_tokens_per_second": 10376.954
    },
    {
      "epoch": 0.34201221472195437,
      "grad_norm": 1.8722063302993774,
      "learning_rate": 1.3272962483829237e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 31351872,
      "step": 266,
      "train_runtime": 3020.55,
      "train_tokens_per_second": 10379.524
    },
    {
      "epoch": 0.343297974927676,
      "grad_norm": 1.941330909729004,
      "learning_rate": 1.3247089262613198e-05,
      "loss": 0.9481,
      "num_input_tokens_seen": 31458432,
      "step": 267,
      "train_runtime": 3029.8631,
      "train_tokens_per_second": 10382.79
    },
    {
      "epoch": 0.34458373513339763,
      "grad_norm": 2.109315872192383,
      "learning_rate": 1.3221216041397157e-05,
      "loss": 1.1085,
      "num_input_tokens_seen": 31566048,
      "step": 268,
      "train_runtime": 3039.2325,
      "train_tokens_per_second": 10386.191
    },
    {
      "epoch": 0.34586949533911926,
      "grad_norm": 1.9429577589035034,
      "learning_rate": 1.3195342820181114e-05,
      "loss": 1.1085,
      "num_input_tokens_seen": 31679008,
      "step": 269,
      "train_runtime": 3049.0419,
      "train_tokens_per_second": 10389.824
    },
    {
      "epoch": 0.3471552555448409,
      "grad_norm": 1.8708595037460327,
      "learning_rate": 1.3169469598965073e-05,
      "loss": 1.0415,
      "num_input_tokens_seen": 31805376,
      "step": 270,
      "train_runtime": 3060.3402,
      "train_tokens_per_second": 10392.758
    },
    {
      "epoch": 0.3484410157505625,
      "grad_norm": 1.890461802482605,
      "learning_rate": 1.314359637774903e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 31887200,
      "step": 271,
      "train_runtime": 3085.3071,
      "train_tokens_per_second": 10335.179
    },
    {
      "epoch": 0.34972677595628415,
      "grad_norm": 1.8630510568618774,
      "learning_rate": 1.3117723156532989e-05,
      "loss": 1.0943,
      "num_input_tokens_seen": 32037536,
      "step": 272,
      "train_runtime": 3098.8904,
      "train_tokens_per_second": 10338.39
    },
    {
      "epoch": 0.3510125361620058,
      "grad_norm": 1.9622751474380493,
      "learning_rate": 1.3091849935316946e-05,
      "loss": 1.0348,
      "num_input_tokens_seen": 32152288,
      "step": 273,
      "train_runtime": 3108.918,
      "train_tokens_per_second": 10341.954
    },
    {
      "epoch": 0.3522982963677274,
      "grad_norm": 1.8532695770263672,
      "learning_rate": 1.3065976714100907e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 32254592,
      "step": 274,
      "train_runtime": 3117.7914,
      "train_tokens_per_second": 10345.333
    },
    {
      "epoch": 0.35358405657344905,
      "grad_norm": 1.8529016971588135,
      "learning_rate": 1.3040103492884866e-05,
      "loss": 1.1095,
      "num_input_tokens_seen": 32377696,
      "step": 275,
      "train_runtime": 3128.8339,
      "train_tokens_per_second": 10348.167
    },
    {
      "epoch": 0.3548698167791707,
      "grad_norm": 1.9654872417449951,
      "learning_rate": 1.3014230271668823e-05,
      "loss": 1.1777,
      "num_input_tokens_seen": 32464224,
      "step": 276,
      "train_runtime": 3136.2376,
      "train_tokens_per_second": 10351.328
    },
    {
      "epoch": 0.3561555769848923,
      "grad_norm": 1.9332945346832275,
      "learning_rate": 1.2988357050452782e-05,
      "loss": 1.0694,
      "num_input_tokens_seen": 32543360,
      "step": 277,
      "train_runtime": 3142.9151,
      "train_tokens_per_second": 10354.515
    },
    {
      "epoch": 0.35744133719061394,
      "grad_norm": 2.014310121536255,
      "learning_rate": 1.2962483829236741e-05,
      "loss": 1.1293,
      "num_input_tokens_seen": 32652000,
      "step": 278,
      "train_runtime": 3152.361,
      "train_tokens_per_second": 10357.951
    },
    {
      "epoch": 0.35872709739633557,
      "grad_norm": 1.9047924280166626,
      "learning_rate": 1.2936610608020698e-05,
      "loss": 1.0027,
      "num_input_tokens_seen": 32764256,
      "step": 279,
      "train_runtime": 3162.2615,
      "train_tokens_per_second": 10361.021
    },
    {
      "epoch": 0.3600128576020572,
      "grad_norm": 1.825059413909912,
      "learning_rate": 1.2910737386804659e-05,
      "loss": 1.1741,
      "num_input_tokens_seen": 32843744,
      "step": 280,
      "train_runtime": 3168.977,
      "train_tokens_per_second": 10364.147
    },
    {
      "epoch": 0.36129861780777883,
      "grad_norm": 1.831803321838379,
      "learning_rate": 1.2884864165588618e-05,
      "loss": 1.022,
      "num_input_tokens_seen": 32964992,
      "step": 281,
      "train_runtime": 3179.9785,
      "train_tokens_per_second": 10366.42
    },
    {
      "epoch": 0.36258437801350046,
      "grad_norm": 1.70236074924469,
      "learning_rate": 1.2858990944372575e-05,
      "loss": 1.0487,
      "num_input_tokens_seen": 33080992,
      "step": 282,
      "train_runtime": 3190.2018,
      "train_tokens_per_second": 10369.561
    },
    {
      "epoch": 0.3638701382192221,
      "grad_norm": 1.796125888824463,
      "learning_rate": 1.2833117723156534e-05,
      "loss": 1.0899,
      "num_input_tokens_seen": 33226368,
      "step": 283,
      "train_runtime": 3203.4864,
      "train_tokens_per_second": 10371.94
    },
    {
      "epoch": 0.3651558984249437,
      "grad_norm": 1.8778120279312134,
      "learning_rate": 1.2807244501940492e-05,
      "loss": 1.0362,
      "num_input_tokens_seen": 33339552,
      "step": 284,
      "train_runtime": 3213.4705,
      "train_tokens_per_second": 10374.937
    },
    {
      "epoch": 0.36644165863066536,
      "grad_norm": 1.868843913078308,
      "learning_rate": 1.278137128072445e-05,
      "loss": 1.0095,
      "num_input_tokens_seen": 33425280,
      "step": 285,
      "train_runtime": 3220.8266,
      "train_tokens_per_second": 10377.858
    },
    {
      "epoch": 0.367727418836387,
      "grad_norm": 1.8552757501602173,
      "learning_rate": 1.275549805950841e-05,
      "loss": 0.9967,
      "num_input_tokens_seen": 33519872,
      "step": 286,
      "train_runtime": 3228.9988,
      "train_tokens_per_second": 10380.887
    },
    {
      "epoch": 0.3690131790421087,
      "grad_norm": 1.8477954864501953,
      "learning_rate": 1.272962483829237e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 33670336,
      "step": 287,
      "train_runtime": 3243.1369,
      "train_tokens_per_second": 10382.027
    },
    {
      "epoch": 0.3702989392478303,
      "grad_norm": 1.7322572469711304,
      "learning_rate": 1.2703751617076327e-05,
      "loss": 1.0259,
      "num_input_tokens_seen": 33777152,
      "step": 288,
      "train_runtime": 3252.4159,
      "train_tokens_per_second": 10385.25
    },
    {
      "epoch": 0.37158469945355194,
      "grad_norm": 1.8179206848144531,
      "learning_rate": 1.2677878395860286e-05,
      "loss": 1.095,
      "num_input_tokens_seen": 33933760,
      "step": 289,
      "train_runtime": 3267.0848,
      "train_tokens_per_second": 10386.556
    },
    {
      "epoch": 0.37287045965927357,
      "grad_norm": 1.927725076675415,
      "learning_rate": 1.2652005174644244e-05,
      "loss": 1.0971,
      "num_input_tokens_seen": 34016064,
      "step": 290,
      "train_runtime": 3274.2027,
      "train_tokens_per_second": 10389.114
    },
    {
      "epoch": 0.3741562198649952,
      "grad_norm": 1.8900138139724731,
      "learning_rate": 1.2626131953428203e-05,
      "loss": 1.0385,
      "num_input_tokens_seen": 34105728,
      "step": 291,
      "train_runtime": 3281.9209,
      "train_tokens_per_second": 10392.002
    },
    {
      "epoch": 0.37544198007071683,
      "grad_norm": 1.8767197132110596,
      "learning_rate": 1.260025873221216e-05,
      "loss": 1.2325,
      "num_input_tokens_seen": 34257888,
      "step": 292,
      "train_runtime": 3296.0865,
      "train_tokens_per_second": 10393.504
    },
    {
      "epoch": 0.37672774027643846,
      "grad_norm": 1.6599918603897095,
      "learning_rate": 1.257438551099612e-05,
      "loss": 1.0944,
      "num_input_tokens_seen": 34370272,
      "step": 293,
      "train_runtime": 3305.9389,
      "train_tokens_per_second": 10396.524
    },
    {
      "epoch": 0.3780135004821601,
      "grad_norm": 1.9018691778182983,
      "learning_rate": 1.254851228978008e-05,
      "loss": 1.1485,
      "num_input_tokens_seen": 34497536,
      "step": 294,
      "train_runtime": 3317.4286,
      "train_tokens_per_second": 10398.878
    },
    {
      "epoch": 0.3792992606878817,
      "grad_norm": 1.9373286962509155,
      "learning_rate": 1.2522639068564037e-05,
      "loss": 1.1207,
      "num_input_tokens_seen": 34655552,
      "step": 295,
      "train_runtime": 3332.4647,
      "train_tokens_per_second": 10399.376
    },
    {
      "epoch": 0.38058502089360335,
      "grad_norm": 1.9912974834442139,
      "learning_rate": 1.2496765847347996e-05,
      "loss": 1.0018,
      "num_input_tokens_seen": 34850176,
      "step": 296,
      "train_runtime": 3351.492,
      "train_tokens_per_second": 10398.406
    },
    {
      "epoch": 0.381870781099325,
      "grad_norm": 2.001157760620117,
      "learning_rate": 1.2470892626131955e-05,
      "loss": 1.0122,
      "num_input_tokens_seen": 34929440,
      "step": 297,
      "train_runtime": 3358.2119,
      "train_tokens_per_second": 10401.202
    },
    {
      "epoch": 0.3831565413050466,
      "grad_norm": 1.9736489057540894,
      "learning_rate": 1.2445019404915912e-05,
      "loss": 1.0287,
      "num_input_tokens_seen": 35043488,
      "step": 298,
      "train_runtime": 3368.304,
      "train_tokens_per_second": 10403.897
    },
    {
      "epoch": 0.38444230151076825,
      "grad_norm": 2.0292022228240967,
      "learning_rate": 1.2419146183699871e-05,
      "loss": 1.1769,
      "num_input_tokens_seen": 35170560,
      "step": 299,
      "train_runtime": 3379.6518,
      "train_tokens_per_second": 10406.563
    },
    {
      "epoch": 0.3857280617164899,
      "grad_norm": 1.7314234972000122,
      "learning_rate": 1.2393272962483832e-05,
      "loss": 0.97,
      "num_input_tokens_seen": 35281056,
      "step": 300,
      "train_runtime": 3389.3149,
      "train_tokens_per_second": 10409.495
    },
    {
      "epoch": 0.3870138219222115,
      "grad_norm": 2.0191800594329834,
      "learning_rate": 1.2367399741267789e-05,
      "loss": 1.1325,
      "num_input_tokens_seen": 35388672,
      "step": 301,
      "train_runtime": 3417.1135,
      "train_tokens_per_second": 10356.306
    },
    {
      "epoch": 0.38829958212793314,
      "grad_norm": 1.8068965673446655,
      "learning_rate": 1.2341526520051748e-05,
      "loss": 1.1311,
      "num_input_tokens_seen": 35523712,
      "step": 302,
      "train_runtime": 3429.3021,
      "train_tokens_per_second": 10358.875
    },
    {
      "epoch": 0.38958534233365477,
      "grad_norm": 1.8851860761642456,
      "learning_rate": 1.2315653298835705e-05,
      "loss": 1.1328,
      "num_input_tokens_seen": 35691712,
      "step": 303,
      "train_runtime": 3445.5901,
      "train_tokens_per_second": 10358.665
    },
    {
      "epoch": 0.3908711025393764,
      "grad_norm": 2.3786840438842773,
      "learning_rate": 1.2289780077619664e-05,
      "loss": 1.1061,
      "num_input_tokens_seen": 35780768,
      "step": 304,
      "train_runtime": 3453.2825,
      "train_tokens_per_second": 10361.379
    },
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 1.793230414390564,
      "learning_rate": 1.2263906856403623e-05,
      "loss": 1.1275,
      "num_input_tokens_seen": 35888608,
      "step": 305,
      "train_runtime": 3462.6599,
      "train_tokens_per_second": 10364.462
    },
    {
      "epoch": 0.39344262295081966,
      "grad_norm": 1.8552781343460083,
      "learning_rate": 1.2238033635187584e-05,
      "loss": 1.1033,
      "num_input_tokens_seen": 36048640,
      "step": 306,
      "train_runtime": 3477.6998,
      "train_tokens_per_second": 10365.656
    },
    {
      "epoch": 0.3947283831565413,
      "grad_norm": 1.8553136587142944,
      "learning_rate": 1.2212160413971541e-05,
      "loss": 1.0171,
      "num_input_tokens_seen": 36150336,
      "step": 307,
      "train_runtime": 3486.8518,
      "train_tokens_per_second": 10367.615
    },
    {
      "epoch": 0.3960141433622629,
      "grad_norm": 1.8456462621688843,
      "learning_rate": 1.21862871927555e-05,
      "loss": 1.0815,
      "num_input_tokens_seen": 36262560,
      "step": 308,
      "train_runtime": 3496.9741,
      "train_tokens_per_second": 10369.696
    },
    {
      "epoch": 0.39729990356798456,
      "grad_norm": 2.0002105236053467,
      "learning_rate": 1.2160413971539457e-05,
      "loss": 1.0209,
      "num_input_tokens_seen": 36364320,
      "step": 309,
      "train_runtime": 3506.0556,
      "train_tokens_per_second": 10371.86
    },
    {
      "epoch": 0.3985856637737062,
      "grad_norm": 1.8822708129882812,
      "learning_rate": 1.2134540750323416e-05,
      "loss": 1.1019,
      "num_input_tokens_seen": 36472480,
      "step": 310,
      "train_runtime": 3515.5179,
      "train_tokens_per_second": 10374.71
    },
    {
      "epoch": 0.3998714239794278,
      "grad_norm": 1.954473614692688,
      "learning_rate": 1.2108667529107373e-05,
      "loss": 1.2242,
      "num_input_tokens_seen": 36585792,
      "step": 311,
      "train_runtime": 3525.5023,
      "train_tokens_per_second": 10377.469
    },
    {
      "epoch": 0.40115718418514945,
      "grad_norm": 1.8884960412979126,
      "learning_rate": 1.2082794307891334e-05,
      "loss": 1.1584,
      "num_input_tokens_seen": 36700448,
      "step": 312,
      "train_runtime": 3536.0058,
      "train_tokens_per_second": 10379.069
    },
    {
      "epoch": 0.40115718418514945,
      "eval_loss": 1.112414836883545,
      "eval_runtime": 22.4354,
      "eval_samples_per_second": 44.305,
      "eval_steps_per_second": 1.426,
      "num_input_tokens_seen": 36700448,
      "step": 312
    },
    {
      "epoch": 0.4024429443908711,
      "grad_norm": 1.7745190858840942,
      "learning_rate": 1.2056921086675293e-05,
      "loss": 1.0609,
      "num_input_tokens_seen": 36819552,
      "step": 313,
      "train_runtime": 3569.0494,
      "train_tokens_per_second": 10316.347
    },
    {
      "epoch": 0.4037287045965927,
      "grad_norm": 1.9588580131530762,
      "learning_rate": 1.2031047865459252e-05,
      "loss": 1.0342,
      "num_input_tokens_seen": 36901792,
      "step": 314,
      "train_runtime": 3576.0188,
      "train_tokens_per_second": 10319.239
    },
    {
      "epoch": 0.40501446480231434,
      "grad_norm": 1.886616826057434,
      "learning_rate": 1.200517464424321e-05,
      "loss": 1.0539,
      "num_input_tokens_seen": 37002112,
      "step": 315,
      "train_runtime": 3584.6465,
      "train_tokens_per_second": 10322.388
    },
    {
      "epoch": 0.406300225008036,
      "grad_norm": 1.9062621593475342,
      "learning_rate": 1.1979301423027168e-05,
      "loss": 1.0985,
      "num_input_tokens_seen": 37111936,
      "step": 316,
      "train_runtime": 3594.3563,
      "train_tokens_per_second": 10325.058
    },
    {
      "epoch": 0.4075859852137576,
      "grad_norm": 2.068682909011841,
      "learning_rate": 1.1953428201811125e-05,
      "loss": 1.097,
      "num_input_tokens_seen": 37218208,
      "step": 317,
      "train_runtime": 3603.7296,
      "train_tokens_per_second": 10327.692
    },
    {
      "epoch": 0.4088717454194793,
      "grad_norm": 1.958736538887024,
      "learning_rate": 1.1927554980595084e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 37345632,
      "step": 318,
      "train_runtime": 3615.4338,
      "train_tokens_per_second": 10329.502
    },
    {
      "epoch": 0.4101575056252009,
      "grad_norm": 1.8023805618286133,
      "learning_rate": 1.1901681759379045e-05,
      "loss": 0.9653,
      "num_input_tokens_seen": 37443360,
      "step": 319,
      "train_runtime": 3623.7748,
      "train_tokens_per_second": 10332.695
    },
    {
      "epoch": 0.41144326583092256,
      "grad_norm": 1.9169106483459473,
      "learning_rate": 1.1875808538163002e-05,
      "loss": 1.0838,
      "num_input_tokens_seen": 37545600,
      "step": 320,
      "train_runtime": 3632.642,
      "train_tokens_per_second": 10335.618
    },
    {
      "epoch": 0.4127290260366442,
      "grad_norm": 1.8038839101791382,
      "learning_rate": 1.1849935316946961e-05,
      "loss": 1.1069,
      "num_input_tokens_seen": 37666080,
      "step": 321,
      "train_runtime": 3643.3493,
      "train_tokens_per_second": 10338.311
    },
    {
      "epoch": 0.4140147862423658,
      "grad_norm": 1.7721644639968872,
      "learning_rate": 1.1824062095730919e-05,
      "loss": 1.0839,
      "num_input_tokens_seen": 37822144,
      "step": 322,
      "train_runtime": 3657.9869,
      "train_tokens_per_second": 10339.606
    },
    {
      "epoch": 0.41530054644808745,
      "grad_norm": 1.9049701690673828,
      "learning_rate": 1.1798188874514877e-05,
      "loss": 1.0799,
      "num_input_tokens_seen": 37966560,
      "step": 323,
      "train_runtime": 3671.372,
      "train_tokens_per_second": 10341.246
    },
    {
      "epoch": 0.4165863066538091,
      "grad_norm": 1.980131983757019,
      "learning_rate": 1.1772315653298836e-05,
      "loss": 1.0307,
      "num_input_tokens_seen": 38062112,
      "step": 324,
      "train_runtime": 3679.5815,
      "train_tokens_per_second": 10344.141
    },
    {
      "epoch": 0.4178720668595307,
      "grad_norm": 1.9756392240524292,
      "learning_rate": 1.1746442432082797e-05,
      "loss": 1.1018,
      "num_input_tokens_seen": 38150848,
      "step": 325,
      "train_runtime": 3687.0852,
      "train_tokens_per_second": 10347.157
    },
    {
      "epoch": 0.41915782706525234,
      "grad_norm": 1.896883249282837,
      "learning_rate": 1.1720569210866754e-05,
      "loss": 1.1311,
      "num_input_tokens_seen": 38241536,
      "step": 326,
      "train_runtime": 3694.6877,
      "train_tokens_per_second": 10350.411
    },
    {
      "epoch": 0.420443587270974,
      "grad_norm": 1.739008903503418,
      "learning_rate": 1.1694695989650713e-05,
      "loss": 1.0084,
      "num_input_tokens_seen": 38318656,
      "step": 327,
      "train_runtime": 3701.2278,
      "train_tokens_per_second": 10352.958
    },
    {
      "epoch": 0.4217293474766956,
      "grad_norm": 1.8083174228668213,
      "learning_rate": 1.166882276843467e-05,
      "loss": 1.123,
      "num_input_tokens_seen": 38417472,
      "step": 328,
      "train_runtime": 3709.7971,
      "train_tokens_per_second": 10355.68
    },
    {
      "epoch": 0.42301510768241724,
      "grad_norm": 1.8054888248443604,
      "learning_rate": 1.164294954721863e-05,
      "loss": 0.9772,
      "num_input_tokens_seen": 38536576,
      "step": 329,
      "train_runtime": 3720.5243,
      "train_tokens_per_second": 10357.835
    },
    {
      "epoch": 0.42430086788813887,
      "grad_norm": 1.8551595211029053,
      "learning_rate": 1.1617076326002587e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 38623488,
      "step": 330,
      "train_runtime": 3727.8728,
      "train_tokens_per_second": 10360.731
    },
    {
      "epoch": 0.4255866280938605,
      "grad_norm": 1.9173765182495117,
      "learning_rate": 1.1591203104786546e-05,
      "loss": 1.0059,
      "num_input_tokens_seen": 38708384,
      "step": 331,
      "train_runtime": 3752.9725,
      "train_tokens_per_second": 10314.06
    },
    {
      "epoch": 0.42687238829958213,
      "grad_norm": 1.9079426527023315,
      "learning_rate": 1.1565329883570506e-05,
      "loss": 1.1054,
      "num_input_tokens_seen": 38843744,
      "step": 332,
      "train_runtime": 3765.4725,
      "train_tokens_per_second": 10315.769
    },
    {
      "epoch": 0.42815814850530376,
      "grad_norm": 1.8765593767166138,
      "learning_rate": 1.1539456662354465e-05,
      "loss": 1.0543,
      "num_input_tokens_seen": 38952768,
      "step": 333,
      "train_runtime": 3774.9835,
      "train_tokens_per_second": 10318.659
    },
    {
      "epoch": 0.4294439087110254,
      "grad_norm": 1.9778969287872314,
      "learning_rate": 1.1513583441138423e-05,
      "loss": 1.0328,
      "num_input_tokens_seen": 39051136,
      "step": 334,
      "train_runtime": 3783.4403,
      "train_tokens_per_second": 10321.594
    },
    {
      "epoch": 0.430729668916747,
      "grad_norm": 1.9157649278640747,
      "learning_rate": 1.1487710219922382e-05,
      "loss": 1.0765,
      "num_input_tokens_seen": 39151392,
      "step": 335,
      "train_runtime": 3792.1825,
      "train_tokens_per_second": 10324.237
    },
    {
      "epoch": 0.43201542912246865,
      "grad_norm": 1.9655895233154297,
      "learning_rate": 1.1461836998706339e-05,
      "loss": 1.0175,
      "num_input_tokens_seen": 39247360,
      "step": 336,
      "train_runtime": 3800.3977,
      "train_tokens_per_second": 10327.172
    },
    {
      "epoch": 0.4333011893281903,
      "grad_norm": 1.8738690614700317,
      "learning_rate": 1.1435963777490298e-05,
      "loss": 1.103,
      "num_input_tokens_seen": 39336896,
      "step": 337,
      "train_runtime": 3808.0625,
      "train_tokens_per_second": 10329.898
    },
    {
      "epoch": 0.4345869495339119,
      "grad_norm": 1.6234264373779297,
      "learning_rate": 1.1410090556274259e-05,
      "loss": 0.8555,
      "num_input_tokens_seen": 39482304,
      "step": 338,
      "train_runtime": 3821.6722,
      "train_tokens_per_second": 10331.159
    },
    {
      "epoch": 0.43587270973963355,
      "grad_norm": 1.8462566137313843,
      "learning_rate": 1.1384217335058216e-05,
      "loss": 1.1103,
      "num_input_tokens_seen": 39575136,
      "step": 339,
      "train_runtime": 3829.6789,
      "train_tokens_per_second": 10333.8
    },
    {
      "epoch": 0.4371584699453552,
      "grad_norm": 1.850907325744629,
      "learning_rate": 1.1358344113842175e-05,
      "loss": 1.0596,
      "num_input_tokens_seen": 39689024,
      "step": 340,
      "train_runtime": 3839.7717,
      "train_tokens_per_second": 10336.298
    },
    {
      "epoch": 0.4384442301510768,
      "grad_norm": 1.863982915878296,
      "learning_rate": 1.1332470892626132e-05,
      "loss": 1.0263,
      "num_input_tokens_seen": 39816832,
      "step": 341,
      "train_runtime": 3851.71,
      "train_tokens_per_second": 10337.443
    },
    {
      "epoch": 0.43972999035679844,
      "grad_norm": 1.933987021446228,
      "learning_rate": 1.1306597671410091e-05,
      "loss": 1.0908,
      "num_input_tokens_seen": 39960768,
      "step": 342,
      "train_runtime": 3865.0922,
      "train_tokens_per_second": 10338.891
    },
    {
      "epoch": 0.44101575056252007,
      "grad_norm": 1.8219382762908936,
      "learning_rate": 1.128072445019405e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 40086912,
      "step": 343,
      "train_runtime": 3876.2397,
      "train_tokens_per_second": 10341.701
    },
    {
      "epoch": 0.4423015107682417,
      "grad_norm": 1.9455715417861938,
      "learning_rate": 1.1254851228978007e-05,
      "loss": 1.1734,
      "num_input_tokens_seen": 40207200,
      "step": 344,
      "train_runtime": 3887.3565,
      "train_tokens_per_second": 10343.07
    },
    {
      "epoch": 0.44358727097396333,
      "grad_norm": 1.81923246383667,
      "learning_rate": 1.1228978007761968e-05,
      "loss": 1.0239,
      "num_input_tokens_seen": 40346560,
      "step": 345,
      "train_runtime": 3900.1116,
      "train_tokens_per_second": 10344.976
    },
    {
      "epoch": 0.44487303117968496,
      "grad_norm": 1.9423332214355469,
      "learning_rate": 1.1203104786545927e-05,
      "loss": 0.9855,
      "num_input_tokens_seen": 40449888,
      "step": 346,
      "train_runtime": 3909.0681,
      "train_tokens_per_second": 10347.706
    },
    {
      "epoch": 0.4461587913854066,
      "grad_norm": 1.9526299238204956,
      "learning_rate": 1.1177231565329884e-05,
      "loss": 1.0271,
      "num_input_tokens_seen": 40586752,
      "step": 347,
      "train_runtime": 3921.8977,
      "train_tokens_per_second": 10348.753
    },
    {
      "epoch": 0.4474445515911283,
      "grad_norm": 1.8065625429153442,
      "learning_rate": 1.1151358344113843e-05,
      "loss": 1.1239,
      "num_input_tokens_seen": 40696608,
      "step": 348,
      "train_runtime": 3931.7547,
      "train_tokens_per_second": 10350.749
    },
    {
      "epoch": 0.4487303117968499,
      "grad_norm": 1.6349153518676758,
      "learning_rate": 1.11254851228978e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 40814656,
      "step": 349,
      "train_runtime": 3942.1447,
      "train_tokens_per_second": 10353.414
    },
    {
      "epoch": 0.45001607200257154,
      "grad_norm": 1.7468831539154053,
      "learning_rate": 1.109961190168176e-05,
      "loss": 1.0172,
      "num_input_tokens_seen": 40974816,
      "step": 350,
      "train_runtime": 3957.5232,
      "train_tokens_per_second": 10353.651
    },
    {
      "epoch": 0.4513018322082932,
      "grad_norm": 1.9717236757278442,
      "learning_rate": 1.107373868046572e-05,
      "loss": 1.1383,
      "num_input_tokens_seen": 41075776,
      "step": 351,
      "train_runtime": 3966.3179,
      "train_tokens_per_second": 10356.148
    },
    {
      "epoch": 0.4525875924140148,
      "grad_norm": 1.715475082397461,
      "learning_rate": 1.1047865459249679e-05,
      "loss": 1.0579,
      "num_input_tokens_seen": 41192352,
      "step": 352,
      "train_runtime": 3976.6366,
      "train_tokens_per_second": 10358.591
    },
    {
      "epoch": 0.45387335261973644,
      "grad_norm": 1.936943769454956,
      "learning_rate": 1.1021992238033636e-05,
      "loss": 1.1545,
      "num_input_tokens_seen": 41296928,
      "step": 353,
      "train_runtime": 3985.9231,
      "train_tokens_per_second": 10360.694
    },
    {
      "epoch": 0.45515911282545807,
      "grad_norm": 1.7825515270233154,
      "learning_rate": 1.0996119016817595e-05,
      "loss": 1.0933,
      "num_input_tokens_seen": 41390336,
      "step": 354,
      "train_runtime": 3993.8858,
      "train_tokens_per_second": 10363.425
    },
    {
      "epoch": 0.4564448730311797,
      "grad_norm": 1.7963905334472656,
      "learning_rate": 1.0970245795601552e-05,
      "loss": 1.0372,
      "num_input_tokens_seen": 41491968,
      "step": 355,
      "train_runtime": 4002.6542,
      "train_tokens_per_second": 10366.114
    },
    {
      "epoch": 0.45773063323690133,
      "grad_norm": 1.836431860923767,
      "learning_rate": 1.0944372574385511e-05,
      "loss": 1.0373,
      "num_input_tokens_seen": 41613888,
      "step": 356,
      "train_runtime": 4013.7789,
      "train_tokens_per_second": 10367.758
    },
    {
      "epoch": 0.45901639344262296,
      "grad_norm": 1.9172090291976929,
      "learning_rate": 1.0918499353169469e-05,
      "loss": 1.0329,
      "num_input_tokens_seen": 41745664,
      "step": 357,
      "train_runtime": 4025.8189,
      "train_tokens_per_second": 10369.484
    },
    {
      "epoch": 0.4603021536483446,
      "grad_norm": 1.8832635879516602,
      "learning_rate": 1.089262613195343e-05,
      "loss": 1.0411,
      "num_input_tokens_seen": 41879264,
      "step": 358,
      "train_runtime": 4038.136,
      "train_tokens_per_second": 10370.939
    },
    {
      "epoch": 0.4615879138540662,
      "grad_norm": 1.8881222009658813,
      "learning_rate": 1.0866752910737388e-05,
      "loss": 1.1016,
      "num_input_tokens_seen": 42019424,
      "step": 359,
      "train_runtime": 4050.808,
      "train_tokens_per_second": 10373.097
    },
    {
      "epoch": 0.46287367405978785,
      "grad_norm": 1.8251198530197144,
      "learning_rate": 1.0840879689521346e-05,
      "loss": 1.0582,
      "num_input_tokens_seen": 42184096,
      "step": 360,
      "train_runtime": 4066.1572,
      "train_tokens_per_second": 10374.438
    },
    {
      "epoch": 0.4641594342655095,
      "grad_norm": 1.8713456392288208,
      "learning_rate": 1.0815006468305305e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 42279968,
      "step": 361,
      "train_runtime": 4091.1662,
      "train_tokens_per_second": 10334.454
    },
    {
      "epoch": 0.4654451944712311,
      "grad_norm": 1.8834689855575562,
      "learning_rate": 1.0789133247089263e-05,
      "loss": 1.1313,
      "num_input_tokens_seen": 42376320,
      "step": 362,
      "train_runtime": 4099.4271,
      "train_tokens_per_second": 10337.132
    },
    {
      "epoch": 0.46673095467695275,
      "grad_norm": 1.9569921493530273,
      "learning_rate": 1.076326002587322e-05,
      "loss": 0.9795,
      "num_input_tokens_seen": 42480736,
      "step": 363,
      "train_runtime": 4108.5761,
      "train_tokens_per_second": 10339.528
    },
    {
      "epoch": 0.4680167148826744,
      "grad_norm": 1.9009162187576294,
      "learning_rate": 1.0737386804657181e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 42588928,
      "step": 364,
      "train_runtime": 4117.995,
      "train_tokens_per_second": 10342.151
    },
    {
      "epoch": 0.469302475088396,
      "grad_norm": 1.9303570985794067,
      "learning_rate": 1.071151358344114e-05,
      "loss": 1.0805,
      "num_input_tokens_seen": 42702336,
      "step": 365,
      "train_runtime": 4127.9119,
      "train_tokens_per_second": 10344.779
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 1.8810940980911255,
      "learning_rate": 1.0685640362225098e-05,
      "loss": 1.0237,
      "num_input_tokens_seen": 42778176,
      "step": 366,
      "train_runtime": 4134.3506,
      "train_tokens_per_second": 10347.012
    },
    {
      "epoch": 0.47187399549983927,
      "grad_norm": 1.7211192846298218,
      "learning_rate": 1.0659767141009057e-05,
      "loss": 0.9568,
      "num_input_tokens_seen": 42865536,
      "step": 367,
      "train_runtime": 4141.8091,
      "train_tokens_per_second": 10349.472
    },
    {
      "epoch": 0.4731597557055609,
      "grad_norm": 1.7768193483352661,
      "learning_rate": 1.0633893919793014e-05,
      "loss": 1.0588,
      "num_input_tokens_seen": 42971392,
      "step": 368,
      "train_runtime": 4151.0681,
      "train_tokens_per_second": 10351.888
    },
    {
      "epoch": 0.47444551591128253,
      "grad_norm": 1.8423272371292114,
      "learning_rate": 1.0608020698576973e-05,
      "loss": 1.0675,
      "num_input_tokens_seen": 43112160,
      "step": 369,
      "train_runtime": 4164.273,
      "train_tokens_per_second": 10352.866
    },
    {
      "epoch": 0.47573127611700416,
      "grad_norm": 1.7994762659072876,
      "learning_rate": 1.0582147477360933e-05,
      "loss": 1.0544,
      "num_input_tokens_seen": 43221728,
      "step": 370,
      "train_runtime": 4173.8176,
      "train_tokens_per_second": 10355.443
    },
    {
      "epoch": 0.4770170363227258,
      "grad_norm": 1.8866075277328491,
      "learning_rate": 1.0556274256144892e-05,
      "loss": 1.1007,
      "num_input_tokens_seen": 43330080,
      "step": 371,
      "train_runtime": 4183.4643,
      "train_tokens_per_second": 10357.464
    },
    {
      "epoch": 0.4783027965284474,
      "grad_norm": 1.8547290563583374,
      "learning_rate": 1.053040103492885e-05,
      "loss": 0.9975,
      "num_input_tokens_seen": 43423392,
      "step": 372,
      "train_runtime": 4191.3728,
      "train_tokens_per_second": 10360.184
    },
    {
      "epoch": 0.47958855673416906,
      "grad_norm": 2.1289970874786377,
      "learning_rate": 1.0504527813712809e-05,
      "loss": 1.0475,
      "num_input_tokens_seen": 43536064,
      "step": 373,
      "train_runtime": 4201.1006,
      "train_tokens_per_second": 10363.014
    },
    {
      "epoch": 0.4808743169398907,
      "grad_norm": 1.9905496835708618,
      "learning_rate": 1.0478654592496766e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 43635072,
      "step": 374,
      "train_runtime": 4209.5792,
      "train_tokens_per_second": 10365.661
    },
    {
      "epoch": 0.4821600771456123,
      "grad_norm": 1.791810393333435,
      "learning_rate": 1.0452781371280725e-05,
      "loss": 1.0144,
      "num_input_tokens_seen": 43724256,
      "step": 375,
      "train_runtime": 4217.1636,
      "train_tokens_per_second": 10368.167
    },
    {
      "epoch": 0.48344583735133395,
      "grad_norm": 1.828808307647705,
      "learning_rate": 1.0426908150064682e-05,
      "loss": 1.1086,
      "num_input_tokens_seen": 43903168,
      "step": 376,
      "train_runtime": 4234.1848,
      "train_tokens_per_second": 10368.742
    },
    {
      "epoch": 0.4847315975570556,
      "grad_norm": 2.0768697261810303,
      "learning_rate": 1.0401034928848643e-05,
      "loss": 1.1689,
      "num_input_tokens_seen": 44021952,
      "step": 377,
      "train_runtime": 4244.5157,
      "train_tokens_per_second": 10371.49
    },
    {
      "epoch": 0.48601735776277727,
      "grad_norm": 1.960158109664917,
      "learning_rate": 1.0375161707632602e-05,
      "loss": 1.061,
      "num_input_tokens_seen": 44108640,
      "step": 378,
      "train_runtime": 4251.8678,
      "train_tokens_per_second": 10373.944
    },
    {
      "epoch": 0.4873031179684989,
      "grad_norm": 1.9121805429458618,
      "learning_rate": 1.034928848641656e-05,
      "loss": 1.0433,
      "num_input_tokens_seen": 44203648,
      "step": 379,
      "train_runtime": 4260.0015,
      "train_tokens_per_second": 10376.44
    },
    {
      "epoch": 0.48858887817422053,
      "grad_norm": 1.8849984407424927,
      "learning_rate": 1.0323415265200518e-05,
      "loss": 1.1013,
      "num_input_tokens_seen": 44291008,
      "step": 380,
      "train_runtime": 4267.4727,
      "train_tokens_per_second": 10378.744
    },
    {
      "epoch": 0.48987463837994216,
      "grad_norm": 1.863187313079834,
      "learning_rate": 1.0297542043984477e-05,
      "loss": 1.028,
      "num_input_tokens_seen": 44377920,
      "step": 381,
      "train_runtime": 4274.8697,
      "train_tokens_per_second": 10381.116
    },
    {
      "epoch": 0.4911603985856638,
      "grad_norm": 1.8333090543746948,
      "learning_rate": 1.0271668822768434e-05,
      "loss": 1.075,
      "num_input_tokens_seen": 44581920,
      "step": 382,
      "train_runtime": 4294.9622,
      "train_tokens_per_second": 10380.049
    },
    {
      "epoch": 0.4924461587913854,
      "grad_norm": 1.873062252998352,
      "learning_rate": 1.0245795601552395e-05,
      "loss": 1.1039,
      "num_input_tokens_seen": 44754688,
      "step": 383,
      "train_runtime": 4311.2726,
      "train_tokens_per_second": 10380.853
    },
    {
      "epoch": 0.49373191899710706,
      "grad_norm": 1.9315266609191895,
      "learning_rate": 1.0219922380336354e-05,
      "loss": 1.145,
      "num_input_tokens_seen": 44880224,
      "step": 384,
      "train_runtime": 4323.0272,
      "train_tokens_per_second": 10381.666
    },
    {
      "epoch": 0.4950176792028287,
      "grad_norm": 1.731623888015747,
      "learning_rate": 1.0194049159120311e-05,
      "loss": 1.0024,
      "num_input_tokens_seen": 45029664,
      "step": 385,
      "train_runtime": 4336.9208,
      "train_tokens_per_second": 10382.865
    },
    {
      "epoch": 0.4963034394085503,
      "grad_norm": 1.618575096130371,
      "learning_rate": 1.016817593790427e-05,
      "loss": 0.9817,
      "num_input_tokens_seen": 45129600,
      "step": 386,
      "train_runtime": 4345.5214,
      "train_tokens_per_second": 10385.313
    },
    {
      "epoch": 0.49758919961427195,
      "grad_norm": 1.8326330184936523,
      "learning_rate": 1.0142302716688227e-05,
      "loss": 1.0646,
      "num_input_tokens_seen": 45219104,
      "step": 387,
      "train_runtime": 4353.1201,
      "train_tokens_per_second": 10387.746
    },
    {
      "epoch": 0.4988749598199936,
      "grad_norm": 1.9410369396209717,
      "learning_rate": 1.0116429495472186e-05,
      "loss": 1.1241,
      "num_input_tokens_seen": 45380032,
      "step": 388,
      "train_runtime": 4368.2134,
      "train_tokens_per_second": 10388.694
    },
    {
      "epoch": 0.5001607200257152,
      "grad_norm": 2.0092506408691406,
      "learning_rate": 1.0090556274256145e-05,
      "loss": 1.0946,
      "num_input_tokens_seen": 45512480,
      "step": 389,
      "train_runtime": 4380.1221,
      "train_tokens_per_second": 10390.687
    },
    {
      "epoch": 0.5014464802314368,
      "grad_norm": 2.0018155574798584,
      "learning_rate": 1.0064683053040106e-05,
      "loss": 1.0753,
      "num_input_tokens_seen": 45640032,
      "step": 390,
      "train_runtime": 4391.8578,
      "train_tokens_per_second": 10391.965
    },
    {
      "epoch": 0.5014464802314368,
      "eval_loss": 1.1071221828460693,
      "eval_runtime": 22.1149,
      "eval_samples_per_second": 44.947,
      "eval_steps_per_second": 1.447,
      "num_input_tokens_seen": 45640032,
      "step": 390
    },
    {
      "epoch": 0.5027322404371585,
      "grad_norm": 1.8650777339935303,
      "learning_rate": 1.0038809831824063e-05,
      "loss": 1.036,
      "num_input_tokens_seen": 45720768,
      "step": 391,
      "train_runtime": 4437.5417,
      "train_tokens_per_second": 10303.175
    },
    {
      "epoch": 0.50401800064288,
      "grad_norm": 1.9369052648544312,
      "learning_rate": 1.0012936610608022e-05,
      "loss": 1.1169,
      "num_input_tokens_seen": 45842240,
      "step": 392,
      "train_runtime": 4448.5568,
      "train_tokens_per_second": 10304.969
    },
    {
      "epoch": 0.5053037608486017,
      "grad_norm": 1.8346235752105713,
      "learning_rate": 9.98706338939198e-06,
      "loss": 1.0749,
      "num_input_tokens_seen": 45918976,
      "step": 393,
      "train_runtime": 4454.9464,
      "train_tokens_per_second": 10307.414
    },
    {
      "epoch": 0.5065895210543234,
      "grad_norm": 2.0440127849578857,
      "learning_rate": 9.961190168175938e-06,
      "loss": 1.0516,
      "num_input_tokens_seen": 45999104,
      "step": 394,
      "train_runtime": 4461.6884,
      "train_tokens_per_second": 10309.798
    },
    {
      "epoch": 0.507875281260045,
      "grad_norm": 3.248145341873169,
      "learning_rate": 9.935316946959897e-06,
      "loss": 1.1728,
      "num_input_tokens_seen": 46154144,
      "step": 395,
      "train_runtime": 4476.3079,
      "train_tokens_per_second": 10310.762
    },
    {
      "epoch": 0.5091610414657667,
      "grad_norm": 1.918129801750183,
      "learning_rate": 9.909443725743856e-06,
      "loss": 1.0521,
      "num_input_tokens_seen": 46325344,
      "step": 396,
      "train_runtime": 4492.8544,
      "train_tokens_per_second": 10310.894
    },
    {
      "epoch": 0.5104468016714883,
      "grad_norm": 1.8980101346969604,
      "learning_rate": 9.883570504527814e-06,
      "loss": 1.0749,
      "num_input_tokens_seen": 46411072,
      "step": 397,
      "train_runtime": 4500.1395,
      "train_tokens_per_second": 10313.252
    },
    {
      "epoch": 0.51173256187721,
      "grad_norm": 1.7899583578109741,
      "learning_rate": 9.857697283311774e-06,
      "loss": 1.0462,
      "num_input_tokens_seen": 46509216,
      "step": 398,
      "train_runtime": 4508.575,
      "train_tokens_per_second": 10315.724
    },
    {
      "epoch": 0.5130183220829315,
      "grad_norm": 1.8656114339828491,
      "learning_rate": 9.831824062095732e-06,
      "loss": 1.0682,
      "num_input_tokens_seen": 46644160,
      "step": 399,
      "train_runtime": 4520.8245,
      "train_tokens_per_second": 10317.622
    },
    {
      "epoch": 0.5143040822886532,
      "grad_norm": 1.847536325454712,
      "learning_rate": 9.80595084087969e-06,
      "loss": 1.0864,
      "num_input_tokens_seen": 46783808,
      "step": 400,
      "train_runtime": 4533.3771,
      "train_tokens_per_second": 10319.858
    },
    {
      "epoch": 0.5155898424943748,
      "grad_norm": 1.8390229940414429,
      "learning_rate": 9.78007761966365e-06,
      "loss": 0.9831,
      "num_input_tokens_seen": 46926240,
      "step": 401,
      "train_runtime": 4546.39,
      "train_tokens_per_second": 10321.649
    },
    {
      "epoch": 0.5168756027000965,
      "grad_norm": 1.8536787033081055,
      "learning_rate": 9.754204398447608e-06,
      "loss": 1.0727,
      "num_input_tokens_seen": 47071040,
      "step": 402,
      "train_runtime": 4559.4688,
      "train_tokens_per_second": 10323.799
    },
    {
      "epoch": 0.518161362905818,
      "grad_norm": 1.8034573793411255,
      "learning_rate": 9.728331177231566e-06,
      "loss": 1.1378,
      "num_input_tokens_seen": 47179744,
      "step": 403,
      "train_runtime": 4569.0897,
      "train_tokens_per_second": 10325.852
    },
    {
      "epoch": 0.5194471231115397,
      "grad_norm": 1.8764989376068115,
      "learning_rate": 9.702457956015525e-06,
      "loss": 1.0454,
      "num_input_tokens_seen": 47293760,
      "step": 404,
      "train_runtime": 4579.0502,
      "train_tokens_per_second": 10328.29
    },
    {
      "epoch": 0.5207328833172613,
      "grad_norm": 1.9220011234283447,
      "learning_rate": 9.676584734799484e-06,
      "loss": 1.1761,
      "num_input_tokens_seen": 47392832,
      "step": 405,
      "train_runtime": 4587.61,
      "train_tokens_per_second": 10330.615
    },
    {
      "epoch": 0.522018643522983,
      "grad_norm": 1.879625916481018,
      "learning_rate": 9.650711513583441e-06,
      "loss": 1.0923,
      "num_input_tokens_seen": 47487712,
      "step": 406,
      "train_runtime": 4595.7576,
      "train_tokens_per_second": 10332.945
    },
    {
      "epoch": 0.5233044037287046,
      "grad_norm": 1.839417815208435,
      "learning_rate": 9.624838292367402e-06,
      "loss": 1.1236,
      "num_input_tokens_seen": 47620096,
      "step": 407,
      "train_runtime": 4607.8581,
      "train_tokens_per_second": 10334.541
    },
    {
      "epoch": 0.5245901639344263,
      "grad_norm": 1.8451459407806396,
      "learning_rate": 9.598965071151359e-06,
      "loss": 1.0462,
      "num_input_tokens_seen": 47700000,
      "step": 408,
      "train_runtime": 4614.6366,
      "train_tokens_per_second": 10336.675
    },
    {
      "epoch": 0.5258759241401478,
      "grad_norm": 1.9718786478042603,
      "learning_rate": 9.573091849935318e-06,
      "loss": 0.9997,
      "num_input_tokens_seen": 47834944,
      "step": 409,
      "train_runtime": 4626.8909,
      "train_tokens_per_second": 10338.464
    },
    {
      "epoch": 0.5271616843458695,
      "grad_norm": 1.9811155796051025,
      "learning_rate": 9.547218628719275e-06,
      "loss": 1.0761,
      "num_input_tokens_seen": 47968160,
      "step": 410,
      "train_runtime": 4639.1336,
      "train_tokens_per_second": 10339.896
    },
    {
      "epoch": 0.5284474445515911,
      "grad_norm": 1.8997541666030884,
      "learning_rate": 9.521345407503236e-06,
      "loss": 1.027,
      "num_input_tokens_seen": 48059648,
      "step": 411,
      "train_runtime": 4646.969,
      "train_tokens_per_second": 10342.149
    },
    {
      "epoch": 0.5297332047573128,
      "grad_norm": 1.8173083066940308,
      "learning_rate": 9.495472186287193e-06,
      "loss": 1.1339,
      "num_input_tokens_seen": 48181504,
      "step": 412,
      "train_runtime": 4658.0106,
      "train_tokens_per_second": 10343.794
    },
    {
      "epoch": 0.5310189649630344,
      "grad_norm": 1.7731293439865112,
      "learning_rate": 9.469598965071152e-06,
      "loss": 0.9738,
      "num_input_tokens_seen": 48280416,
      "step": 413,
      "train_runtime": 4666.5826,
      "train_tokens_per_second": 10345.99
    },
    {
      "epoch": 0.532304725168756,
      "grad_norm": 1.9140669107437134,
      "learning_rate": 9.443725743855111e-06,
      "loss": 1.0163,
      "num_input_tokens_seen": 48409600,
      "step": 414,
      "train_runtime": 4678.3897,
      "train_tokens_per_second": 10347.492
    },
    {
      "epoch": 0.5335904853744776,
      "grad_norm": 1.872811198234558,
      "learning_rate": 9.41785252263907e-06,
      "loss": 0.9948,
      "num_input_tokens_seen": 48530944,
      "step": 415,
      "train_runtime": 4689.1516,
      "train_tokens_per_second": 10349.622
    },
    {
      "epoch": 0.5348762455801993,
      "grad_norm": 1.7188026905059814,
      "learning_rate": 9.391979301423027e-06,
      "loss": 1.0364,
      "num_input_tokens_seen": 48621344,
      "step": 416,
      "train_runtime": 4696.9538,
      "train_tokens_per_second": 10351.676
    },
    {
      "epoch": 0.5361620057859209,
      "grad_norm": 1.9433099031448364,
      "learning_rate": 9.366106080206988e-06,
      "loss": 1.0985,
      "num_input_tokens_seen": 48730336,
      "step": 417,
      "train_runtime": 4706.5206,
      "train_tokens_per_second": 10353.792
    },
    {
      "epoch": 0.5374477659916426,
      "grad_norm": 1.8120436668395996,
      "learning_rate": 9.340232858990945e-06,
      "loss": 1.0375,
      "num_input_tokens_seen": 48876416,
      "step": 418,
      "train_runtime": 4720.1751,
      "train_tokens_per_second": 10354.789
    },
    {
      "epoch": 0.5387335261973641,
      "grad_norm": 1.8917372226715088,
      "learning_rate": 9.314359637774904e-06,
      "loss": 1.0179,
      "num_input_tokens_seen": 48968384,
      "step": 419,
      "train_runtime": 4728.029,
      "train_tokens_per_second": 10357.04
    },
    {
      "epoch": 0.5400192864030858,
      "grad_norm": 1.9676231145858765,
      "learning_rate": 9.288486416558863e-06,
      "loss": 1.0241,
      "num_input_tokens_seen": 49076928,
      "step": 420,
      "train_runtime": 4737.4729,
      "train_tokens_per_second": 10359.305
    },
    {
      "epoch": 0.5413050466088074,
      "grad_norm": 1.799108862876892,
      "learning_rate": 9.262613195342822e-06,
      "loss": 1.1027,
      "num_input_tokens_seen": 49202080,
      "step": 421,
      "train_runtime": 4765.3676,
      "train_tokens_per_second": 10324.928
    },
    {
      "epoch": 0.5425908068145291,
      "grad_norm": 1.6937819719314575,
      "learning_rate": 9.23673997412678e-06,
      "loss": 1.0448,
      "num_input_tokens_seen": 49279744,
      "step": 422,
      "train_runtime": 4772.0012,
      "train_tokens_per_second": 10326.851
    },
    {
      "epoch": 0.5438765670202508,
      "grad_norm": 1.8214465379714966,
      "learning_rate": 9.210866752910738e-06,
      "loss": 1.0483,
      "num_input_tokens_seen": 49426304,
      "step": 423,
      "train_runtime": 4785.0768,
      "train_tokens_per_second": 10329.26
    },
    {
      "epoch": 0.5451623272259724,
      "grad_norm": 2.017838954925537,
      "learning_rate": 9.184993531694697e-06,
      "loss": 1.0529,
      "num_input_tokens_seen": 49552544,
      "step": 424,
      "train_runtime": 4796.8995,
      "train_tokens_per_second": 10330.119
    },
    {
      "epoch": 0.546448087431694,
      "grad_norm": 1.8154163360595703,
      "learning_rate": 9.159120310478656e-06,
      "loss": 1.0638,
      "num_input_tokens_seen": 49681152,
      "step": 425,
      "train_runtime": 4808.5011,
      "train_tokens_per_second": 10331.941
    },
    {
      "epoch": 0.5477338476374156,
      "grad_norm": 2.039503574371338,
      "learning_rate": 9.133247089262613e-06,
      "loss": 1.1204,
      "num_input_tokens_seen": 49810272,
      "step": 426,
      "train_runtime": 4820.4338,
      "train_tokens_per_second": 10333.151
    },
    {
      "epoch": 0.5490196078431373,
      "grad_norm": 1.8751721382141113,
      "learning_rate": 9.107373868046572e-06,
      "loss": 1.0905,
      "num_input_tokens_seen": 49931552,
      "step": 427,
      "train_runtime": 4831.6448,
      "train_tokens_per_second": 10334.276
    },
    {
      "epoch": 0.5503053680488589,
      "grad_norm": 1.9049893617630005,
      "learning_rate": 9.081500646830531e-06,
      "loss": 1.0296,
      "num_input_tokens_seen": 50054880,
      "step": 428,
      "train_runtime": 4842.7281,
      "train_tokens_per_second": 10336.091
    },
    {
      "epoch": 0.5515911282545806,
      "grad_norm": 1.743492603302002,
      "learning_rate": 9.055627425614489e-06,
      "loss": 1.0613,
      "num_input_tokens_seen": 50135776,
      "step": 429,
      "train_runtime": 4849.6827,
      "train_tokens_per_second": 10337.95
    },
    {
      "epoch": 0.5528768884603021,
      "grad_norm": 1.752612590789795,
      "learning_rate": 9.02975420439845e-06,
      "loss": 1.0087,
      "num_input_tokens_seen": 50233920,
      "step": 430,
      "train_runtime": 4858.229,
      "train_tokens_per_second": 10339.965
    },
    {
      "epoch": 0.5541626486660238,
      "grad_norm": 1.793007493019104,
      "learning_rate": 9.003880983182406e-06,
      "loss": 1.0386,
      "num_input_tokens_seen": 50364480,
      "step": 431,
      "train_runtime": 4870.2509,
      "train_tokens_per_second": 10341.249
    },
    {
      "epoch": 0.5554484088717454,
      "grad_norm": 1.8837916851043701,
      "learning_rate": 8.978007761966365e-06,
      "loss": 1.0343,
      "num_input_tokens_seen": 50477568,
      "step": 432,
      "train_runtime": 4880.2122,
      "train_tokens_per_second": 10343.314
    },
    {
      "epoch": 0.5567341690774671,
      "grad_norm": 1.8399786949157715,
      "learning_rate": 8.952134540750324e-06,
      "loss": 1.0474,
      "num_input_tokens_seen": 50587904,
      "step": 433,
      "train_runtime": 4889.8391,
      "train_tokens_per_second": 10345.515
    },
    {
      "epoch": 0.5580199292831887,
      "grad_norm": 1.9835869073867798,
      "learning_rate": 8.926261319534283e-06,
      "loss": 1.0986,
      "num_input_tokens_seen": 50748288,
      "step": 434,
      "train_runtime": 4904.8056,
      "train_tokens_per_second": 10346.646
    },
    {
      "epoch": 0.5593056894889104,
      "grad_norm": 2.022273540496826,
      "learning_rate": 8.90038809831824e-06,
      "loss": 1.078,
      "num_input_tokens_seen": 50829376,
      "step": 435,
      "train_runtime": 4911.6238,
      "train_tokens_per_second": 10348.793
    },
    {
      "epoch": 0.5605914496946319,
      "grad_norm": 1.9517446756362915,
      "learning_rate": 8.874514877102201e-06,
      "loss": 1.0425,
      "num_input_tokens_seen": 50990240,
      "step": 436,
      "train_runtime": 4927.2726,
      "train_tokens_per_second": 10348.573
    },
    {
      "epoch": 0.5618772099003536,
      "grad_norm": 1.8184734582901,
      "learning_rate": 8.848641655886159e-06,
      "loss": 1.201,
      "num_input_tokens_seen": 51092736,
      "step": 437,
      "train_runtime": 4936.1299,
      "train_tokens_per_second": 10350.768
    },
    {
      "epoch": 0.5631629701060752,
      "grad_norm": 1.876457929611206,
      "learning_rate": 8.822768434670118e-06,
      "loss": 1.0678,
      "num_input_tokens_seen": 51179680,
      "step": 438,
      "train_runtime": 4943.6088,
      "train_tokens_per_second": 10352.696
    },
    {
      "epoch": 0.5644487303117969,
      "grad_norm": 1.8654677867889404,
      "learning_rate": 8.796895213454075e-06,
      "loss": 1.0735,
      "num_input_tokens_seen": 51292160,
      "step": 439,
      "train_runtime": 4953.7338,
      "train_tokens_per_second": 10354.242
    },
    {
      "epoch": 0.5657344905175185,
      "grad_norm": 1.7864034175872803,
      "learning_rate": 8.771021992238035e-06,
      "loss": 1.0005,
      "num_input_tokens_seen": 51402400,
      "step": 440,
      "train_runtime": 4963.8198,
      "train_tokens_per_second": 10355.412
    },
    {
      "epoch": 0.5670202507232401,
      "grad_norm": 1.8480334281921387,
      "learning_rate": 8.745148771021993e-06,
      "loss": 1.146,
      "num_input_tokens_seen": 51503840,
      "step": 441,
      "train_runtime": 4972.6909,
      "train_tokens_per_second": 10357.338
    },
    {
      "epoch": 0.5683060109289617,
      "grad_norm": 1.8786932229995728,
      "learning_rate": 8.719275549805952e-06,
      "loss": 1.032,
      "num_input_tokens_seen": 51625920,
      "step": 442,
      "train_runtime": 4983.5486,
      "train_tokens_per_second": 10359.269
    },
    {
      "epoch": 0.5695917711346834,
      "grad_norm": 1.8805335760116577,
      "learning_rate": 8.69340232858991e-06,
      "loss": 1.0604,
      "num_input_tokens_seen": 51766848,
      "step": 443,
      "train_runtime": 4996.6032,
      "train_tokens_per_second": 10360.408
    },
    {
      "epoch": 0.570877531340405,
      "grad_norm": 2.044856548309326,
      "learning_rate": 8.66752910737387e-06,
      "loss": 1.0327,
      "num_input_tokens_seen": 51885568,
      "step": 444,
      "train_runtime": 5007.1763,
      "train_tokens_per_second": 10362.241
    },
    {
      "epoch": 0.5721632915461267,
      "grad_norm": 1.7294365167617798,
      "learning_rate": 8.641655886157827e-06,
      "loss": 1.0273,
      "num_input_tokens_seen": 51987968,
      "step": 445,
      "train_runtime": 5016.2418,
      "train_tokens_per_second": 10363.928
    },
    {
      "epoch": 0.5734490517518482,
      "grad_norm": 1.9049499034881592,
      "learning_rate": 8.615782664941786e-06,
      "loss": 1.0087,
      "num_input_tokens_seen": 52075232,
      "step": 446,
      "train_runtime": 5023.7352,
      "train_tokens_per_second": 10365.839
    },
    {
      "epoch": 0.5747348119575699,
      "grad_norm": 1.995016098022461,
      "learning_rate": 8.589909443725745e-06,
      "loss": 1.082,
      "num_input_tokens_seen": 52181120,
      "step": 447,
      "train_runtime": 5033.0217,
      "train_tokens_per_second": 10367.752
    },
    {
      "epoch": 0.5760205721632915,
      "grad_norm": 1.8061517477035522,
      "learning_rate": 8.564036222509704e-06,
      "loss": 1.0338,
      "num_input_tokens_seen": 52312864,
      "step": 448,
      "train_runtime": 5045.4751,
      "train_tokens_per_second": 10368.273
    },
    {
      "epoch": 0.5773063323690132,
      "grad_norm": 1.906706690788269,
      "learning_rate": 8.538163001293663e-06,
      "loss": 1.0313,
      "num_input_tokens_seen": 52492128,
      "step": 449,
      "train_runtime": 5062.8483,
      "train_tokens_per_second": 10368.102
    },
    {
      "epoch": 0.5785920925747348,
      "grad_norm": 1.9607821702957153,
      "learning_rate": 8.51228978007762e-06,
      "loss": 1.0598,
      "num_input_tokens_seen": 52580800,
      "step": 450,
      "train_runtime": 5070.3956,
      "train_tokens_per_second": 10370.157
    },
    {
      "epoch": 0.5798778527804564,
      "grad_norm": 1.8769546747207642,
      "learning_rate": 8.486416558861579e-06,
      "loss": 1.0721,
      "num_input_tokens_seen": 52715552,
      "step": 451,
      "train_runtime": 5099.7454,
      "train_tokens_per_second": 10336.899
    },
    {
      "epoch": 0.581163612986178,
      "grad_norm": 1.8132247924804688,
      "learning_rate": 8.460543337645536e-06,
      "loss": 1.0979,
      "num_input_tokens_seen": 52884896,
      "step": 452,
      "train_runtime": 5115.8206,
      "train_tokens_per_second": 10337.52
    },
    {
      "epoch": 0.5824493731918997,
      "grad_norm": 1.8635915517807007,
      "learning_rate": 8.434670116429497e-06,
      "loss": 1.0368,
      "num_input_tokens_seen": 53008224,
      "step": 453,
      "train_runtime": 5126.8881,
      "train_tokens_per_second": 10339.259
    },
    {
      "epoch": 0.5837351333976214,
      "grad_norm": 1.903838038444519,
      "learning_rate": 8.408796895213454e-06,
      "loss": 0.9926,
      "num_input_tokens_seen": 53141920,
      "step": 454,
      "train_runtime": 5138.9508,
      "train_tokens_per_second": 10341.006
    },
    {
      "epoch": 0.585020893603343,
      "grad_norm": 1.836340069770813,
      "learning_rate": 8.382923673997413e-06,
      "loss": 1.0946,
      "num_input_tokens_seen": 53272960,
      "step": 455,
      "train_runtime": 5150.7368,
      "train_tokens_per_second": 10342.784
    },
    {
      "epoch": 0.5863066538090647,
      "grad_norm": 1.8196651935577393,
      "learning_rate": 8.357050452781372e-06,
      "loss": 1.0399,
      "num_input_tokens_seen": 53416992,
      "step": 456,
      "train_runtime": 5163.847,
      "train_tokens_per_second": 10344.418
    },
    {
      "epoch": 0.5875924140147862,
      "grad_norm": 1.8723372220993042,
      "learning_rate": 8.331177231565331e-06,
      "loss": 1.1476,
      "num_input_tokens_seen": 53533632,
      "step": 457,
      "train_runtime": 5174.1711,
      "train_tokens_per_second": 10346.32
    },
    {
      "epoch": 0.5888781742205079,
      "grad_norm": 1.8454405069351196,
      "learning_rate": 8.305304010349288e-06,
      "loss": 1.031,
      "num_input_tokens_seen": 53631552,
      "step": 458,
      "train_runtime": 5182.534,
      "train_tokens_per_second": 10348.519
    },
    {
      "epoch": 0.5901639344262295,
      "grad_norm": 1.9296767711639404,
      "learning_rate": 8.279430789133249e-06,
      "loss": 1.036,
      "num_input_tokens_seen": 53754528,
      "step": 459,
      "train_runtime": 5193.4642,
      "train_tokens_per_second": 10350.419
    },
    {
      "epoch": 0.5914496946319512,
      "grad_norm": 1.8463411331176758,
      "learning_rate": 8.253557567917206e-06,
      "loss": 0.9773,
      "num_input_tokens_seen": 53828992,
      "step": 460,
      "train_runtime": 5199.7389,
      "train_tokens_per_second": 10352.249
    },
    {
      "epoch": 0.5927354548376728,
      "grad_norm": 1.9723373651504517,
      "learning_rate": 8.227684346701165e-06,
      "loss": 1.1172,
      "num_input_tokens_seen": 54010208,
      "step": 461,
      "train_runtime": 5217.2951,
      "train_tokens_per_second": 10352.147
    },
    {
      "epoch": 0.5940212150433944,
      "grad_norm": 1.9189200401306152,
      "learning_rate": 8.201811125485124e-06,
      "loss": 1.0317,
      "num_input_tokens_seen": 54134656,
      "step": 462,
      "train_runtime": 5228.207,
      "train_tokens_per_second": 10354.344
    },
    {
      "epoch": 0.595306975249116,
      "grad_norm": 1.6208829879760742,
      "learning_rate": 8.175937904269083e-06,
      "loss": 0.9256,
      "num_input_tokens_seen": 54268224,
      "step": 463,
      "train_runtime": 5240.6612,
      "train_tokens_per_second": 10355.225
    },
    {
      "epoch": 0.5965927354548377,
      "grad_norm": 1.9756633043289185,
      "learning_rate": 8.15006468305304e-06,
      "loss": 1.0466,
      "num_input_tokens_seen": 54384544,
      "step": 464,
      "train_runtime": 5251.3101,
      "train_tokens_per_second": 10356.376
    },
    {
      "epoch": 0.5978784956605593,
      "grad_norm": 1.8915356397628784,
      "learning_rate": 8.124191461837e-06,
      "loss": 1.0826,
      "num_input_tokens_seen": 54546208,
      "step": 465,
      "train_runtime": 5266.7741,
      "train_tokens_per_second": 10356.664
    },
    {
      "epoch": 0.599164255866281,
      "grad_norm": 1.9341920614242554,
      "learning_rate": 8.098318240620958e-06,
      "loss": 1.0257,
      "num_input_tokens_seen": 54644864,
      "step": 466,
      "train_runtime": 5275.3751,
      "train_tokens_per_second": 10358.479
    },
    {
      "epoch": 0.6004500160720025,
      "grad_norm": 1.772900104522705,
      "learning_rate": 8.072445019404917e-06,
      "loss": 1.0155,
      "num_input_tokens_seen": 54775808,
      "step": 467,
      "train_runtime": 5287.3482,
      "train_tokens_per_second": 10359.788
    },
    {
      "epoch": 0.6017357762777242,
      "grad_norm": 1.8051958084106445,
      "learning_rate": 8.046571798188875e-06,
      "loss": 1.0296,
      "num_input_tokens_seen": 54865024,
      "step": 468,
      "train_runtime": 5294.9344,
      "train_tokens_per_second": 10361.795
    },
    {
      "epoch": 0.6017357762777242,
      "eval_loss": 1.104156732559204,
      "eval_runtime": 22.4988,
      "eval_samples_per_second": 44.18,
      "eval_steps_per_second": 1.422,
      "num_input_tokens_seen": 54865024,
      "step": 468
    },
    {
      "epoch": 0.6030215364834458,
      "grad_norm": 1.9409384727478027,
      "learning_rate": 8.020698576972833e-06,
      "loss": 1.0859,
      "num_input_tokens_seen": 55004192,
      "step": 469,
      "train_runtime": 5330.358,
      "train_tokens_per_second": 10319.043
    },
    {
      "epoch": 0.6043072966891675,
      "grad_norm": 1.7463427782058716,
      "learning_rate": 7.994825355756792e-06,
      "loss": 1.0851,
      "num_input_tokens_seen": 55115104,
      "step": 470,
      "train_runtime": 5340.0483,
      "train_tokens_per_second": 10321.087
    },
    {
      "epoch": 0.6055930568948891,
      "grad_norm": 1.958573579788208,
      "learning_rate": 7.96895213454075e-06,
      "loss": 1.0175,
      "num_input_tokens_seen": 55232576,
      "step": 471,
      "train_runtime": 5350.618,
      "train_tokens_per_second": 10322.654
    },
    {
      "epoch": 0.6068788171006108,
      "grad_norm": 1.8370342254638672,
      "learning_rate": 7.94307891332471e-06,
      "loss": 1.0919,
      "num_input_tokens_seen": 55327584,
      "step": 472,
      "train_runtime": 5358.7331,
      "train_tokens_per_second": 10324.751
    },
    {
      "epoch": 0.6081645773063323,
      "grad_norm": 1.8849748373031616,
      "learning_rate": 7.917205692108668e-06,
      "loss": 1.061,
      "num_input_tokens_seen": 55460224,
      "step": 473,
      "train_runtime": 5371.0369,
      "train_tokens_per_second": 10325.795
    },
    {
      "epoch": 0.609450337512054,
      "grad_norm": 1.8478233814239502,
      "learning_rate": 7.891332470892627e-06,
      "loss": 1.0702,
      "num_input_tokens_seen": 55603200,
      "step": 474,
      "train_runtime": 5384.0612,
      "train_tokens_per_second": 10327.371
    },
    {
      "epoch": 0.6107360977177756,
      "grad_norm": 1.7144081592559814,
      "learning_rate": 7.865459249676586e-06,
      "loss": 1.0623,
      "num_input_tokens_seen": 55715008,
      "step": 475,
      "train_runtime": 5393.8386,
      "train_tokens_per_second": 10329.38
    },
    {
      "epoch": 0.6120218579234973,
      "grad_norm": 1.895750641822815,
      "learning_rate": 7.839586028460545e-06,
      "loss": 1.0541,
      "num_input_tokens_seen": 55814400,
      "step": 476,
      "train_runtime": 5402.4312,
      "train_tokens_per_second": 10331.349
    },
    {
      "epoch": 0.6133076181292189,
      "grad_norm": 1.997480034828186,
      "learning_rate": 7.813712807244502e-06,
      "loss": 1.1358,
      "num_input_tokens_seen": 55905248,
      "step": 477,
      "train_runtime": 5410.2074,
      "train_tokens_per_second": 10333.291
    },
    {
      "epoch": 0.6145933783349405,
      "grad_norm": 1.9574437141418457,
      "learning_rate": 7.787839586028462e-06,
      "loss": 1.0872,
      "num_input_tokens_seen": 55996224,
      "step": 478,
      "train_runtime": 5418.1887,
      "train_tokens_per_second": 10334.86
    },
    {
      "epoch": 0.6158791385406621,
      "grad_norm": 1.738667607307434,
      "learning_rate": 7.76196636481242e-06,
      "loss": 1.0266,
      "num_input_tokens_seen": 56089024,
      "step": 479,
      "train_runtime": 5426.2457,
      "train_tokens_per_second": 10336.617
    },
    {
      "epoch": 0.6171648987463838,
      "grad_norm": 1.8634562492370605,
      "learning_rate": 7.736093143596379e-06,
      "loss": 1.1175,
      "num_input_tokens_seen": 56203232,
      "step": 480,
      "train_runtime": 5436.2029,
      "train_tokens_per_second": 10338.693
    },
    {
      "epoch": 0.6184506589521054,
      "grad_norm": 1.8316950798034668,
      "learning_rate": 7.710219922380336e-06,
      "loss": 0.9945,
      "num_input_tokens_seen": 56293504,
      "step": 481,
      "train_runtime": 5460.5999,
      "train_tokens_per_second": 10309.033
    },
    {
      "epoch": 0.6197364191578271,
      "grad_norm": 1.9369782209396362,
      "learning_rate": 7.684346701164297e-06,
      "loss": 1.082,
      "num_input_tokens_seen": 56402752,
      "step": 482,
      "train_runtime": 5470.187,
      "train_tokens_per_second": 10310.937
    },
    {
      "epoch": 0.6210221793635486,
      "grad_norm": 1.7741248607635498,
      "learning_rate": 7.658473479948254e-06,
      "loss": 1.0357,
      "num_input_tokens_seen": 56505376,
      "step": 483,
      "train_runtime": 5479.036,
      "train_tokens_per_second": 10313.014
    },
    {
      "epoch": 0.6223079395692703,
      "grad_norm": 1.9275310039520264,
      "learning_rate": 7.632600258732213e-06,
      "loss": 1.0627,
      "num_input_tokens_seen": 56607264,
      "step": 484,
      "train_runtime": 5487.9536,
      "train_tokens_per_second": 10314.822
    },
    {
      "epoch": 0.623593699774992,
      "grad_norm": 1.8197015523910522,
      "learning_rate": 7.606727037516172e-06,
      "loss": 0.966,
      "num_input_tokens_seen": 56736576,
      "step": 485,
      "train_runtime": 5500.0845,
      "train_tokens_per_second": 10315.583
    },
    {
      "epoch": 0.6248794599807136,
      "grad_norm": 1.7420237064361572,
      "learning_rate": 7.58085381630013e-06,
      "loss": 1.0898,
      "num_input_tokens_seen": 56832608,
      "step": 486,
      "train_runtime": 5508.3187,
      "train_tokens_per_second": 10317.596
    },
    {
      "epoch": 0.6261652201864353,
      "grad_norm": 1.7904372215270996,
      "learning_rate": 7.554980595084088e-06,
      "loss": 0.9849,
      "num_input_tokens_seen": 56918528,
      "step": 487,
      "train_runtime": 5515.7042,
      "train_tokens_per_second": 10319.358
    },
    {
      "epoch": 0.6274509803921569,
      "grad_norm": 1.901534914970398,
      "learning_rate": 7.529107373868048e-06,
      "loss": 1.0978,
      "num_input_tokens_seen": 57060512,
      "step": 488,
      "train_runtime": 5528.8699,
      "train_tokens_per_second": 10320.466
    },
    {
      "epoch": 0.6287367405978785,
      "grad_norm": 1.8489160537719727,
      "learning_rate": 7.503234152652006e-06,
      "loss": 1.0536,
      "num_input_tokens_seen": 57153408,
      "step": 489,
      "train_runtime": 5536.9185,
      "train_tokens_per_second": 10322.241
    },
    {
      "epoch": 0.6300225008036001,
      "grad_norm": 1.8837651014328003,
      "learning_rate": 7.477360931435964e-06,
      "loss": 1.0979,
      "num_input_tokens_seen": 57244704,
      "step": 490,
      "train_runtime": 5544.8307,
      "train_tokens_per_second": 10323.977
    },
    {
      "epoch": 0.6313082610093218,
      "grad_norm": 1.9377018213272095,
      "learning_rate": 7.451487710219923e-06,
      "loss": 0.9867,
      "num_input_tokens_seen": 57382048,
      "step": 491,
      "train_runtime": 5557.3194,
      "train_tokens_per_second": 10325.49
    },
    {
      "epoch": 0.6325940212150434,
      "grad_norm": 1.971827507019043,
      "learning_rate": 7.425614489003882e-06,
      "loss": 1.1096,
      "num_input_tokens_seen": 57526912,
      "step": 492,
      "train_runtime": 5570.7926,
      "train_tokens_per_second": 10326.522
    },
    {
      "epoch": 0.6338797814207651,
      "grad_norm": 1.8177952766418457,
      "learning_rate": 7.39974126778784e-06,
      "loss": 1.0804,
      "num_input_tokens_seen": 57651648,
      "step": 493,
      "train_runtime": 5582.2541,
      "train_tokens_per_second": 10327.665
    },
    {
      "epoch": 0.6351655416264866,
      "grad_norm": 1.9560736417770386,
      "learning_rate": 7.373868046571799e-06,
      "loss": 1.0818,
      "num_input_tokens_seen": 57742848,
      "step": 494,
      "train_runtime": 5590.0563,
      "train_tokens_per_second": 10329.565
    },
    {
      "epoch": 0.6364513018322083,
      "grad_norm": 2.095134735107422,
      "learning_rate": 7.347994825355757e-06,
      "loss": 1.051,
      "num_input_tokens_seen": 57848384,
      "step": 495,
      "train_runtime": 5599.5099,
      "train_tokens_per_second": 10330.973
    },
    {
      "epoch": 0.6377370620379299,
      "grad_norm": 2.0076119899749756,
      "learning_rate": 7.322121604139716e-06,
      "loss": 1.1306,
      "num_input_tokens_seen": 57996032,
      "step": 496,
      "train_runtime": 5613.6125,
      "train_tokens_per_second": 10331.321
    },
    {
      "epoch": 0.6390228222436516,
      "grad_norm": 1.7557566165924072,
      "learning_rate": 7.296248382923674e-06,
      "loss": 0.9975,
      "num_input_tokens_seen": 58088800,
      "step": 497,
      "train_runtime": 5621.5772,
      "train_tokens_per_second": 10333.185
    },
    {
      "epoch": 0.6403085824493732,
      "grad_norm": 1.983188271522522,
      "learning_rate": 7.270375161707633e-06,
      "loss": 0.9731,
      "num_input_tokens_seen": 58211040,
      "step": 498,
      "train_runtime": 5632.8912,
      "train_tokens_per_second": 10334.132
    },
    {
      "epoch": 0.6415943426550949,
      "grad_norm": 1.955208659172058,
      "learning_rate": 7.244501940491591e-06,
      "loss": 1.0637,
      "num_input_tokens_seen": 58334432,
      "step": 499,
      "train_runtime": 5644.1895,
      "train_tokens_per_second": 10335.307
    },
    {
      "epoch": 0.6428801028608164,
      "grad_norm": 1.839125156402588,
      "learning_rate": 7.21862871927555e-06,
      "loss": 1.0653,
      "num_input_tokens_seen": 58444576,
      "step": 500,
      "train_runtime": 5653.6395,
      "train_tokens_per_second": 10337.514
    },
    {
      "epoch": 0.6441658630665381,
      "grad_norm": 1.8079696893692017,
      "learning_rate": 7.192755498059509e-06,
      "loss": 1.0928,
      "num_input_tokens_seen": 58603360,
      "step": 501,
      "train_runtime": 5668.8043,
      "train_tokens_per_second": 10337.87
    },
    {
      "epoch": 0.6454516232722597,
      "grad_norm": 1.7496858835220337,
      "learning_rate": 7.166882276843467e-06,
      "loss": 0.9713,
      "num_input_tokens_seen": 58771264,
      "step": 502,
      "train_runtime": 5684.7051,
      "train_tokens_per_second": 10338.49
    },
    {
      "epoch": 0.6467373834779814,
      "grad_norm": 2.015773057937622,
      "learning_rate": 7.1410090556274255e-06,
      "loss": 1.0992,
      "num_input_tokens_seen": 58931392,
      "step": 503,
      "train_runtime": 5699.5505,
      "train_tokens_per_second": 10339.656
    },
    {
      "epoch": 0.648023143683703,
      "grad_norm": 1.7812045812606812,
      "learning_rate": 7.115135834411385e-06,
      "loss": 1.0177,
      "num_input_tokens_seen": 59031904,
      "step": 504,
      "train_runtime": 5708.5567,
      "train_tokens_per_second": 10340.951
    },
    {
      "epoch": 0.6493089038894246,
      "grad_norm": 2.226012945175171,
      "learning_rate": 7.089262613195343e-06,
      "loss": 1.0775,
      "num_input_tokens_seen": 59170880,
      "step": 505,
      "train_runtime": 5721.3851,
      "train_tokens_per_second": 10342.055
    },
    {
      "epoch": 0.6505946640951462,
      "grad_norm": 1.7196251153945923,
      "learning_rate": 7.0633893919793015e-06,
      "loss": 1.0867,
      "num_input_tokens_seen": 59305920,
      "step": 506,
      "train_runtime": 5733.791,
      "train_tokens_per_second": 10343.23
    },
    {
      "epoch": 0.6518804243008679,
      "grad_norm": 2.0985634326934814,
      "learning_rate": 7.037516170763261e-06,
      "loss": 1.0544,
      "num_input_tokens_seen": 59429280,
      "step": 507,
      "train_runtime": 5744.9517,
      "train_tokens_per_second": 10344.609
    },
    {
      "epoch": 0.6531661845065895,
      "grad_norm": 1.983805775642395,
      "learning_rate": 7.0116429495472195e-06,
      "loss": 0.9989,
      "num_input_tokens_seen": 59511424,
      "step": 508,
      "train_runtime": 5751.8269,
      "train_tokens_per_second": 10346.525
    },
    {
      "epoch": 0.6544519447123112,
      "grad_norm": 1.8357391357421875,
      "learning_rate": 6.985769728331178e-06,
      "loss": 1.0282,
      "num_input_tokens_seen": 59629280,
      "step": 509,
      "train_runtime": 5762.4785,
      "train_tokens_per_second": 10347.853
    },
    {
      "epoch": 0.6557377049180327,
      "grad_norm": 1.8154447078704834,
      "learning_rate": 6.959896507115136e-06,
      "loss": 1.0676,
      "num_input_tokens_seen": 59728704,
      "step": 510,
      "train_runtime": 5771.0809,
      "train_tokens_per_second": 10349.656
    },
    {
      "epoch": 0.6570234651237544,
      "grad_norm": 1.966147541999817,
      "learning_rate": 6.9340232858990955e-06,
      "loss": 1.1222,
      "num_input_tokens_seen": 59858304,
      "step": 511,
      "train_runtime": 5799.4165,
      "train_tokens_per_second": 10321.436
    },
    {
      "epoch": 0.658309225329476,
      "grad_norm": 1.942854404449463,
      "learning_rate": 6.908150064683054e-06,
      "loss": 1.1369,
      "num_input_tokens_seen": 59962496,
      "step": 512,
      "train_runtime": 5808.3201,
      "train_tokens_per_second": 10323.552
    },
    {
      "epoch": 0.6595949855351977,
      "grad_norm": 2.0717549324035645,
      "learning_rate": 6.882276843467012e-06,
      "loss": 1.0623,
      "num_input_tokens_seen": 60104320,
      "step": 513,
      "train_runtime": 5821.4241,
      "train_tokens_per_second": 10324.676
    },
    {
      "epoch": 0.6608807457409194,
      "grad_norm": 2.0552477836608887,
      "learning_rate": 6.856403622250971e-06,
      "loss": 1.079,
      "num_input_tokens_seen": 60239488,
      "step": 514,
      "train_runtime": 5833.6761,
      "train_tokens_per_second": 10326.163
    },
    {
      "epoch": 0.662166505946641,
      "grad_norm": 1.8785006999969482,
      "learning_rate": 6.83053040103493e-06,
      "loss": 1.0287,
      "num_input_tokens_seen": 60372960,
      "step": 515,
      "train_runtime": 5846.0146,
      "train_tokens_per_second": 10327.2
    },
    {
      "epoch": 0.6634522661523626,
      "grad_norm": 2.1240034103393555,
      "learning_rate": 6.804657179818888e-06,
      "loss": 1.093,
      "num_input_tokens_seen": 60524480,
      "step": 516,
      "train_runtime": 5860.475,
      "train_tokens_per_second": 10327.572
    },
    {
      "epoch": 0.6647380263580842,
      "grad_norm": 1.815077304840088,
      "learning_rate": 6.778783958602847e-06,
      "loss": 1.0346,
      "num_input_tokens_seen": 60655744,
      "step": 517,
      "train_runtime": 5872.4212,
      "train_tokens_per_second": 10328.916
    },
    {
      "epoch": 0.6660237865638059,
      "grad_norm": 1.8126624822616577,
      "learning_rate": 6.752910737386805e-06,
      "loss": 1.0734,
      "num_input_tokens_seen": 60770176,
      "step": 518,
      "train_runtime": 5882.5093,
      "train_tokens_per_second": 10330.655
    },
    {
      "epoch": 0.6673095467695275,
      "grad_norm": 1.7896870374679565,
      "learning_rate": 6.727037516170764e-06,
      "loss": 1.0592,
      "num_input_tokens_seen": 60889088,
      "step": 519,
      "train_runtime": 5892.9745,
      "train_tokens_per_second": 10332.488
    },
    {
      "epoch": 0.6685953069752492,
      "grad_norm": 1.8001877069473267,
      "learning_rate": 6.701164294954723e-06,
      "loss": 0.9968,
      "num_input_tokens_seen": 61065792,
      "step": 520,
      "train_runtime": 5909.8573,
      "train_tokens_per_second": 10332.871
    },
    {
      "epoch": 0.6698810671809707,
      "grad_norm": 1.8801597356796265,
      "learning_rate": 6.675291073738681e-06,
      "loss": 1.0645,
      "num_input_tokens_seen": 61235808,
      "step": 521,
      "train_runtime": 5926.3712,
      "train_tokens_per_second": 10332.766
    },
    {
      "epoch": 0.6711668273866924,
      "grad_norm": 1.8804895877838135,
      "learning_rate": 6.649417852522639e-06,
      "loss": 1.0528,
      "num_input_tokens_seen": 61393632,
      "step": 522,
      "train_runtime": 5940.9242,
      "train_tokens_per_second": 10334.02
    },
    {
      "epoch": 0.672452587592414,
      "grad_norm": 1.9801836013793945,
      "learning_rate": 6.623544631306599e-06,
      "loss": 1.1185,
      "num_input_tokens_seen": 61529632,
      "step": 523,
      "train_runtime": 5953.2464,
      "train_tokens_per_second": 10335.475
    },
    {
      "epoch": 0.6737383477981357,
      "grad_norm": 1.8925362825393677,
      "learning_rate": 6.597671410090557e-06,
      "loss": 1.1512,
      "num_input_tokens_seen": 61631840,
      "step": 524,
      "train_runtime": 5962.0027,
      "train_tokens_per_second": 10337.439
    },
    {
      "epoch": 0.6750241080038573,
      "grad_norm": 1.7096664905548096,
      "learning_rate": 6.571798188874515e-06,
      "loss": 1.0007,
      "num_input_tokens_seen": 61719392,
      "step": 525,
      "train_runtime": 5969.4912,
      "train_tokens_per_second": 10339.138
    },
    {
      "epoch": 0.676309868209579,
      "grad_norm": 1.7562854290008545,
      "learning_rate": 6.545924967658473e-06,
      "loss": 0.9808,
      "num_input_tokens_seen": 61904448,
      "step": 526,
      "train_runtime": 5987.4689,
      "train_tokens_per_second": 10339.001
    },
    {
      "epoch": 0.6775956284153005,
      "grad_norm": 1.9827765226364136,
      "learning_rate": 6.520051746442433e-06,
      "loss": 1.0306,
      "num_input_tokens_seen": 62045888,
      "step": 527,
      "train_runtime": 6000.3698,
      "train_tokens_per_second": 10340.344
    },
    {
      "epoch": 0.6788813886210222,
      "grad_norm": 1.9799909591674805,
      "learning_rate": 6.494178525226391e-06,
      "loss": 0.9475,
      "num_input_tokens_seen": 62144736,
      "step": 528,
      "train_runtime": 6008.9407,
      "train_tokens_per_second": 10342.045
    },
    {
      "epoch": 0.6801671488267438,
      "grad_norm": 1.928503155708313,
      "learning_rate": 6.468305304010349e-06,
      "loss": 0.973,
      "num_input_tokens_seen": 62266400,
      "step": 529,
      "train_runtime": 6019.7454,
      "train_tokens_per_second": 10343.693
    },
    {
      "epoch": 0.6814529090324655,
      "grad_norm": 1.9949895143508911,
      "learning_rate": 6.442432082794309e-06,
      "loss": 1.1023,
      "num_input_tokens_seen": 62377888,
      "step": 530,
      "train_runtime": 6029.3488,
      "train_tokens_per_second": 10345.709
    },
    {
      "epoch": 0.682738669238187,
      "grad_norm": 1.8765671253204346,
      "learning_rate": 6.416558861578267e-06,
      "loss": 1.0263,
      "num_input_tokens_seen": 62462016,
      "step": 531,
      "train_runtime": 6036.392,
      "train_tokens_per_second": 10347.574
    },
    {
      "epoch": 0.6840244294439087,
      "grad_norm": 1.9741531610488892,
      "learning_rate": 6.390685640362225e-06,
      "loss": 1.2246,
      "num_input_tokens_seen": 62610176,
      "step": 532,
      "train_runtime": 6050.1111,
      "train_tokens_per_second": 10348.599
    },
    {
      "epoch": 0.6853101896496303,
      "grad_norm": 1.8716306686401367,
      "learning_rate": 6.364812419146185e-06,
      "loss": 1.0364,
      "num_input_tokens_seen": 62709280,
      "step": 533,
      "train_runtime": 6058.5723,
      "train_tokens_per_second": 10350.505
    },
    {
      "epoch": 0.686595949855352,
      "grad_norm": 1.8738210201263428,
      "learning_rate": 6.338939197930143e-06,
      "loss": 1.0923,
      "num_input_tokens_seen": 62874912,
      "step": 534,
      "train_runtime": 6074.0138,
      "train_tokens_per_second": 10351.46
    },
    {
      "epoch": 0.6878817100610736,
      "grad_norm": 1.8222072124481201,
      "learning_rate": 6.313065976714101e-06,
      "loss": 1.0699,
      "num_input_tokens_seen": 62994208,
      "step": 535,
      "train_runtime": 6084.6192,
      "train_tokens_per_second": 10353.024
    },
    {
      "epoch": 0.6891674702667953,
      "grad_norm": 1.8478925228118896,
      "learning_rate": 6.28719275549806e-06,
      "loss": 1.1806,
      "num_input_tokens_seen": 63145056,
      "step": 536,
      "train_runtime": 6098.4851,
      "train_tokens_per_second": 10354.22
    },
    {
      "epoch": 0.6904532304725168,
      "grad_norm": 2.016186475753784,
      "learning_rate": 6.261319534282018e-06,
      "loss": 1.0304,
      "num_input_tokens_seen": 63269120,
      "step": 537,
      "train_runtime": 6110.0516,
      "train_tokens_per_second": 10354.924
    },
    {
      "epoch": 0.6917389906782385,
      "grad_norm": 1.8224762678146362,
      "learning_rate": 6.235446313065977e-06,
      "loss": 1.0067,
      "num_input_tokens_seen": 63415744,
      "step": 538,
      "train_runtime": 6123.7782,
      "train_tokens_per_second": 10355.657
    },
    {
      "epoch": 0.6930247508839601,
      "grad_norm": 1.7332342863082886,
      "learning_rate": 6.2095730918499354e-06,
      "loss": 1.0274,
      "num_input_tokens_seen": 63508160,
      "step": 539,
      "train_runtime": 6131.7391,
      "train_tokens_per_second": 10357.283
    },
    {
      "epoch": 0.6943105110896818,
      "grad_norm": 1.7827410697937012,
      "learning_rate": 6.183699870633894e-06,
      "loss": 1.0193,
      "num_input_tokens_seen": 63661440,
      "step": 540,
      "train_runtime": 6146.5269,
      "train_tokens_per_second": 10357.303
    },
    {
      "epoch": 0.6955962712954034,
      "grad_norm": 2.020627498626709,
      "learning_rate": 6.1578266494178525e-06,
      "loss": 1.0964,
      "num_input_tokens_seen": 63780032,
      "step": 541,
      "train_runtime": 6173.4887,
      "train_tokens_per_second": 10331.279
    },
    {
      "epoch": 0.696882031501125,
      "grad_norm": 1.7501956224441528,
      "learning_rate": 6.1319534282018115e-06,
      "loss": 1.0793,
      "num_input_tokens_seen": 63918880,
      "step": 542,
      "train_runtime": 6185.9809,
      "train_tokens_per_second": 10332.861
    },
    {
      "epoch": 0.6981677917068466,
      "grad_norm": 1.8000164031982422,
      "learning_rate": 6.1060802069857704e-06,
      "loss": 0.9915,
      "num_input_tokens_seen": 64021696,
      "step": 543,
      "train_runtime": 6194.7919,
      "train_tokens_per_second": 10334.761
    },
    {
      "epoch": 0.6994535519125683,
      "grad_norm": 1.771580696105957,
      "learning_rate": 6.0802069857697286e-06,
      "loss": 1.0623,
      "num_input_tokens_seen": 64140384,
      "step": 544,
      "train_runtime": 6205.2559,
      "train_tokens_per_second": 10336.461
    },
    {
      "epoch": 0.70073931211829,
      "grad_norm": 1.9515384435653687,
      "learning_rate": 6.054333764553687e-06,
      "loss": 1.1481,
      "num_input_tokens_seen": 64242624,
      "step": 545,
      "train_runtime": 6214.1214,
      "train_tokens_per_second": 10338.167
    },
    {
      "epoch": 0.7020250723240116,
      "grad_norm": 1.8691986799240112,
      "learning_rate": 6.0284605433376465e-06,
      "loss": 1.0209,
      "num_input_tokens_seen": 64314496,
      "step": 546,
      "train_runtime": 6220.1913,
      "train_tokens_per_second": 10339.633
    },
    {
      "epoch": 0.7020250723240116,
      "eval_loss": 1.102497935295105,
      "eval_runtime": 22.3274,
      "eval_samples_per_second": 44.519,
      "eval_steps_per_second": 1.433,
      "num_input_tokens_seen": 64314496,
      "step": 546
    },
    {
      "epoch": 0.7033108325297333,
      "grad_norm": 1.691506266593933,
      "learning_rate": 6.002587322121605e-06,
      "loss": 0.9789,
      "num_input_tokens_seen": 64429248,
      "step": 547,
      "train_runtime": 6252.5897,
      "train_tokens_per_second": 10304.41
    },
    {
      "epoch": 0.7045965927354548,
      "grad_norm": 1.7860418558120728,
      "learning_rate": 5.976714100905563e-06,
      "loss": 1.0317,
      "num_input_tokens_seen": 64559776,
      "step": 548,
      "train_runtime": 6264.281,
      "train_tokens_per_second": 10306.015
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 2.1169979572296143,
      "learning_rate": 5.9508408796895225e-06,
      "loss": 1.0426,
      "num_input_tokens_seen": 64683264,
      "step": 549,
      "train_runtime": 6275.2924,
      "train_tokens_per_second": 10307.61
    },
    {
      "epoch": 0.7071681131468981,
      "grad_norm": 1.937103271484375,
      "learning_rate": 5.924967658473481e-06,
      "loss": 1.0845,
      "num_input_tokens_seen": 64821728,
      "step": 550,
      "train_runtime": 6288.0189,
      "train_tokens_per_second": 10308.768
    },
    {
      "epoch": 0.7084538733526198,
      "grad_norm": 1.8114924430847168,
      "learning_rate": 5.899094437257439e-06,
      "loss": 1.0205,
      "num_input_tokens_seen": 64920384,
      "step": 551,
      "train_runtime": 6296.4281,
      "train_tokens_per_second": 10310.669
    },
    {
      "epoch": 0.7097396335583414,
      "grad_norm": 1.9015802145004272,
      "learning_rate": 5.8732212160413986e-06,
      "loss": 1.0589,
      "num_input_tokens_seen": 65041088,
      "step": 552,
      "train_runtime": 6307.1492,
      "train_tokens_per_second": 10312.28
    },
    {
      "epoch": 0.711025393764063,
      "grad_norm": 1.871267318725586,
      "learning_rate": 5.847347994825357e-06,
      "loss": 1.0599,
      "num_input_tokens_seen": 65166368,
      "step": 553,
      "train_runtime": 6318.2983,
      "train_tokens_per_second": 10313.911
    },
    {
      "epoch": 0.7123111539697846,
      "grad_norm": 1.7762296199798584,
      "learning_rate": 5.821474773609315e-06,
      "loss": 1.1277,
      "num_input_tokens_seen": 65256192,
      "step": 554,
      "train_runtime": 6326.0044,
      "train_tokens_per_second": 10315.546
    },
    {
      "epoch": 0.7135969141755063,
      "grad_norm": 2.015854835510254,
      "learning_rate": 5.795601552393273e-06,
      "loss": 1.048,
      "num_input_tokens_seen": 65379488,
      "step": 555,
      "train_runtime": 6337.3792,
      "train_tokens_per_second": 10316.487
    },
    {
      "epoch": 0.7148826743812279,
      "grad_norm": 1.9349653720855713,
      "learning_rate": 5.769728331177233e-06,
      "loss": 1.0893,
      "num_input_tokens_seen": 65530368,
      "step": 556,
      "train_runtime": 6351.5436,
      "train_tokens_per_second": 10317.235
    },
    {
      "epoch": 0.7161684345869496,
      "grad_norm": 1.8240134716033936,
      "learning_rate": 5.743855109961191e-06,
      "loss": 1.0671,
      "num_input_tokens_seen": 65613952,
      "step": 557,
      "train_runtime": 6358.6261,
      "train_tokens_per_second": 10318.888
    },
    {
      "epoch": 0.7174541947926711,
      "grad_norm": 1.8752096891403198,
      "learning_rate": 5.717981888745149e-06,
      "loss": 1.0451,
      "num_input_tokens_seen": 65725792,
      "step": 558,
      "train_runtime": 6368.3372,
      "train_tokens_per_second": 10320.715
    },
    {
      "epoch": 0.7187399549983928,
      "grad_norm": 1.9915884733200073,
      "learning_rate": 5.692108667529108e-06,
      "loss": 1.0574,
      "num_input_tokens_seen": 65807680,
      "step": 559,
      "train_runtime": 6375.3518,
      "train_tokens_per_second": 10322.204
    },
    {
      "epoch": 0.7200257152041144,
      "grad_norm": 1.7847318649291992,
      "learning_rate": 5.666235446313066e-06,
      "loss": 1.061,
      "num_input_tokens_seen": 65905312,
      "step": 560,
      "train_runtime": 6383.8117,
      "train_tokens_per_second": 10323.818
    },
    {
      "epoch": 0.7213114754098361,
      "grad_norm": 1.7851864099502563,
      "learning_rate": 5.640362225097025e-06,
      "loss": 0.9781,
      "num_input_tokens_seen": 66025056,
      "step": 561,
      "train_runtime": 6394.6288,
      "train_tokens_per_second": 10325.08
    },
    {
      "epoch": 0.7225972356155577,
      "grad_norm": 2.038956642150879,
      "learning_rate": 5.614489003880984e-06,
      "loss": 1.1079,
      "num_input_tokens_seen": 66161536,
      "step": 562,
      "train_runtime": 6406.936,
      "train_tokens_per_second": 10326.549
    },
    {
      "epoch": 0.7238829958212794,
      "grad_norm": 1.8547343015670776,
      "learning_rate": 5.588615782664942e-06,
      "loss": 0.9582,
      "num_input_tokens_seen": 66301920,
      "step": 563,
      "train_runtime": 6419.8653,
      "train_tokens_per_second": 10327.619
    },
    {
      "epoch": 0.7251687560270009,
      "grad_norm": 1.8734216690063477,
      "learning_rate": 5.5627425614489e-06,
      "loss": 1.1257,
      "num_input_tokens_seen": 66416448,
      "step": 564,
      "train_runtime": 6430.0471,
      "train_tokens_per_second": 10329.076
    },
    {
      "epoch": 0.7264545162327226,
      "grad_norm": 1.9347343444824219,
      "learning_rate": 5.53686934023286e-06,
      "loss": 1.0665,
      "num_input_tokens_seen": 66504544,
      "step": 565,
      "train_runtime": 6437.5545,
      "train_tokens_per_second": 10330.715
    },
    {
      "epoch": 0.7277402764384442,
      "grad_norm": 1.8810735940933228,
      "learning_rate": 5.510996119016818e-06,
      "loss": 1.0375,
      "num_input_tokens_seen": 66656448,
      "step": 566,
      "train_runtime": 6451.7561,
      "train_tokens_per_second": 10331.52
    },
    {
      "epoch": 0.7290260366441659,
      "grad_norm": 1.837829351425171,
      "learning_rate": 5.485122897800776e-06,
      "loss": 1.0713,
      "num_input_tokens_seen": 66759904,
      "step": 567,
      "train_runtime": 6460.6494,
      "train_tokens_per_second": 10333.312
    },
    {
      "epoch": 0.7303117968498875,
      "grad_norm": 1.9980813264846802,
      "learning_rate": 5.459249676584734e-06,
      "loss": 0.9986,
      "num_input_tokens_seen": 66894848,
      "step": 568,
      "train_runtime": 6473.0889,
      "train_tokens_per_second": 10334.301
    },
    {
      "epoch": 0.7315975570556091,
      "grad_norm": 1.976579189300537,
      "learning_rate": 5.433376455368694e-06,
      "loss": 1.1012,
      "num_input_tokens_seen": 66996128,
      "step": 569,
      "train_runtime": 6481.9025,
      "train_tokens_per_second": 10335.874
    },
    {
      "epoch": 0.7328833172613307,
      "grad_norm": 1.903847336769104,
      "learning_rate": 5.407503234152652e-06,
      "loss": 1.0675,
      "num_input_tokens_seen": 67141920,
      "step": 570,
      "train_runtime": 6495.5236,
      "train_tokens_per_second": 10336.645
    },
    {
      "epoch": 0.7341690774670524,
      "grad_norm": 1.9221888780593872,
      "learning_rate": 5.38163001293661e-06,
      "loss": 1.0781,
      "num_input_tokens_seen": 67290336,
      "step": 571,
      "train_runtime": 6525.8926,
      "train_tokens_per_second": 10311.285
    },
    {
      "epoch": 0.735454837672774,
      "grad_norm": 1.808961033821106,
      "learning_rate": 5.35575679172057e-06,
      "loss": 1.1462,
      "num_input_tokens_seen": 67395424,
      "step": 572,
      "train_runtime": 6534.8507,
      "train_tokens_per_second": 10313.231
    },
    {
      "epoch": 0.7367405978784957,
      "grad_norm": 1.8847907781600952,
      "learning_rate": 5.329883570504528e-06,
      "loss": 1.0736,
      "num_input_tokens_seen": 67521888,
      "step": 573,
      "train_runtime": 6545.9199,
      "train_tokens_per_second": 10315.111
    },
    {
      "epoch": 0.7380263580842173,
      "grad_norm": 1.8549025058746338,
      "learning_rate": 5.304010349288486e-06,
      "loss": 1.0503,
      "num_input_tokens_seen": 67665568,
      "step": 574,
      "train_runtime": 6559.1726,
      "train_tokens_per_second": 10316.174
    },
    {
      "epoch": 0.7393121182899389,
      "grad_norm": 1.9307861328125,
      "learning_rate": 5.278137128072446e-06,
      "loss": 1.1438,
      "num_input_tokens_seen": 67774688,
      "step": 575,
      "train_runtime": 6568.6043,
      "train_tokens_per_second": 10317.974
    },
    {
      "epoch": 0.7405978784956606,
      "grad_norm": 1.9152367115020752,
      "learning_rate": 5.252263906856404e-06,
      "loss": 1.0097,
      "num_input_tokens_seen": 67861824,
      "step": 576,
      "train_runtime": 6575.9557,
      "train_tokens_per_second": 10319.69
    },
    {
      "epoch": 0.7418836387013822,
      "grad_norm": 1.8602268695831299,
      "learning_rate": 5.2263906856403625e-06,
      "loss": 1.0663,
      "num_input_tokens_seen": 67982592,
      "step": 577,
      "train_runtime": 6586.8236,
      "train_tokens_per_second": 10320.998
    },
    {
      "epoch": 0.7431693989071039,
      "grad_norm": 1.928527593612671,
      "learning_rate": 5.200517464424321e-06,
      "loss": 1.1119,
      "num_input_tokens_seen": 68088704,
      "step": 578,
      "train_runtime": 6595.9873,
      "train_tokens_per_second": 10322.746
    },
    {
      "epoch": 0.7444551591128254,
      "grad_norm": 2.0047895908355713,
      "learning_rate": 5.17464424320828e-06,
      "loss": 1.0698,
      "num_input_tokens_seen": 68230048,
      "step": 579,
      "train_runtime": 6608.8637,
      "train_tokens_per_second": 10324.021
    },
    {
      "epoch": 0.7457409193185471,
      "grad_norm": 1.8268404006958008,
      "learning_rate": 5.1487710219922385e-06,
      "loss": 1.0993,
      "num_input_tokens_seen": 68349472,
      "step": 580,
      "train_runtime": 6619.3705,
      "train_tokens_per_second": 10325.676
    },
    {
      "epoch": 0.7470266795242687,
      "grad_norm": 2.0684757232666016,
      "learning_rate": 5.1228978007761975e-06,
      "loss": 1.0588,
      "num_input_tokens_seen": 68470720,
      "step": 581,
      "train_runtime": 6630.3909,
      "train_tokens_per_second": 10326.8
    },
    {
      "epoch": 0.7483124397299904,
      "grad_norm": 1.9927403926849365,
      "learning_rate": 5.0970245795601556e-06,
      "loss": 1.0661,
      "num_input_tokens_seen": 68583616,
      "step": 582,
      "train_runtime": 6640.5867,
      "train_tokens_per_second": 10327.945
    },
    {
      "epoch": 0.749598199935712,
      "grad_norm": 1.9514095783233643,
      "learning_rate": 5.071151358344114e-06,
      "loss": 1.0833,
      "num_input_tokens_seen": 68700480,
      "step": 583,
      "train_runtime": 6650.9406,
      "train_tokens_per_second": 10329.438
    },
    {
      "epoch": 0.7508839601414337,
      "grad_norm": 2.018437623977661,
      "learning_rate": 5.045278137128073e-06,
      "loss": 1.0668,
      "num_input_tokens_seen": 68822336,
      "step": 584,
      "train_runtime": 6661.8349,
      "train_tokens_per_second": 10330.838
    },
    {
      "epoch": 0.7521697203471552,
      "grad_norm": 1.8875172138214111,
      "learning_rate": 5.019404915912032e-06,
      "loss": 1.0462,
      "num_input_tokens_seen": 68951936,
      "step": 585,
      "train_runtime": 6673.1759,
      "train_tokens_per_second": 10332.702
    },
    {
      "epoch": 0.7534554805528769,
      "grad_norm": 1.818596363067627,
      "learning_rate": 4.99353169469599e-06,
      "loss": 1.04,
      "num_input_tokens_seen": 69135360,
      "step": 586,
      "train_runtime": 6690.4234,
      "train_tokens_per_second": 10333.48
    },
    {
      "epoch": 0.7547412407585985,
      "grad_norm": 1.750368595123291,
      "learning_rate": 4.967658473479949e-06,
      "loss": 1.0268,
      "num_input_tokens_seen": 69267744,
      "step": 587,
      "train_runtime": 6702.7098,
      "train_tokens_per_second": 10334.29
    },
    {
      "epoch": 0.7560270009643202,
      "grad_norm": 1.8207074403762817,
      "learning_rate": 4.941785252263907e-06,
      "loss": 1.0989,
      "num_input_tokens_seen": 69370016,
      "step": 588,
      "train_runtime": 6711.6974,
      "train_tokens_per_second": 10335.689
    },
    {
      "epoch": 0.7573127611700418,
      "grad_norm": 1.9229310750961304,
      "learning_rate": 4.915912031047866e-06,
      "loss": 1.1032,
      "num_input_tokens_seen": 69495744,
      "step": 589,
      "train_runtime": 6722.8508,
      "train_tokens_per_second": 10337.243
    },
    {
      "epoch": 0.7585985213757634,
      "grad_norm": 1.8761484622955322,
      "learning_rate": 4.890038809831825e-06,
      "loss": 1.0723,
      "num_input_tokens_seen": 69637824,
      "step": 590,
      "train_runtime": 6735.7596,
      "train_tokens_per_second": 10338.526
    },
    {
      "epoch": 0.759884281581485,
      "grad_norm": 1.9209434986114502,
      "learning_rate": 4.864165588615783e-06,
      "loss": 1.0539,
      "num_input_tokens_seen": 69746688,
      "step": 591,
      "train_runtime": 6745.3898,
      "train_tokens_per_second": 10339.905
    },
    {
      "epoch": 0.7611700417872067,
      "grad_norm": 1.8482396602630615,
      "learning_rate": 4.838292367399742e-06,
      "loss": 1.1362,
      "num_input_tokens_seen": 69896288,
      "step": 592,
      "train_runtime": 6759.169,
      "train_tokens_per_second": 10340.959
    },
    {
      "epoch": 0.7624558019929283,
      "grad_norm": 1.7221297025680542,
      "learning_rate": 4.812419146183701e-06,
      "loss": 1.0768,
      "num_input_tokens_seen": 69981248,
      "step": 593,
      "train_runtime": 6766.3944,
      "train_tokens_per_second": 10342.473
    },
    {
      "epoch": 0.76374156219865,
      "grad_norm": 1.803056240081787,
      "learning_rate": 4.786545924967659e-06,
      "loss": 1.0445,
      "num_input_tokens_seen": 70077920,
      "step": 594,
      "train_runtime": 6774.6332,
      "train_tokens_per_second": 10344.165
    },
    {
      "epoch": 0.7650273224043715,
      "grad_norm": 2.036080837249756,
      "learning_rate": 4.760672703751618e-06,
      "loss": 1.0569,
      "num_input_tokens_seen": 70207136,
      "step": 595,
      "train_runtime": 6786.3397,
      "train_tokens_per_second": 10345.361
    },
    {
      "epoch": 0.7663130826100932,
      "grad_norm": 1.8578526973724365,
      "learning_rate": 4.734799482535576e-06,
      "loss": 1.0575,
      "num_input_tokens_seen": 70295488,
      "step": 596,
      "train_runtime": 6793.831,
      "train_tokens_per_second": 10346.959
    },
    {
      "epoch": 0.7675988428158148,
      "grad_norm": 1.8725582361221313,
      "learning_rate": 4.708926261319535e-06,
      "loss": 1.0409,
      "num_input_tokens_seen": 70387808,
      "step": 597,
      "train_runtime": 6801.6342,
      "train_tokens_per_second": 10348.661
    },
    {
      "epoch": 0.7688846030215365,
      "grad_norm": 1.8751484155654907,
      "learning_rate": 4.683053040103494e-06,
      "loss": 1.0569,
      "num_input_tokens_seen": 70502912,
      "step": 598,
      "train_runtime": 6811.7789,
      "train_tokens_per_second": 10350.147
    },
    {
      "epoch": 0.7701703632272581,
      "grad_norm": 1.832022786140442,
      "learning_rate": 4.657179818887452e-06,
      "loss": 1.0852,
      "num_input_tokens_seen": 70629696,
      "step": 599,
      "train_runtime": 6822.8214,
      "train_tokens_per_second": 10351.978
    },
    {
      "epoch": 0.7714561234329798,
      "grad_norm": 1.9740808010101318,
      "learning_rate": 4.631306597671411e-06,
      "loss": 1.0614,
      "num_input_tokens_seen": 70732640,
      "step": 600,
      "train_runtime": 6831.678,
      "train_tokens_per_second": 10353.626
    },
    {
      "epoch": 0.7727418836387013,
      "grad_norm": 1.9888503551483154,
      "learning_rate": 4.605433376455369e-06,
      "loss": 1.1406,
      "num_input_tokens_seen": 70851840,
      "step": 601,
      "train_runtime": 6859.0779,
      "train_tokens_per_second": 10329.645
    },
    {
      "epoch": 0.774027643844423,
      "grad_norm": 2.011823892593384,
      "learning_rate": 4.579560155239328e-06,
      "loss": 1.0582,
      "num_input_tokens_seen": 70977120,
      "step": 602,
      "train_runtime": 6870.6102,
      "train_tokens_per_second": 10330.541
    },
    {
      "epoch": 0.7753134040501446,
      "grad_norm": 1.8113346099853516,
      "learning_rate": 4.553686934023286e-06,
      "loss": 1.0753,
      "num_input_tokens_seen": 71089760,
      "step": 603,
      "train_runtime": 6880.2562,
      "train_tokens_per_second": 10332.429
    },
    {
      "epoch": 0.7765991642558663,
      "grad_norm": 1.8886075019836426,
      "learning_rate": 4.527813712807244e-06,
      "loss": 1.067,
      "num_input_tokens_seen": 71200000,
      "step": 604,
      "train_runtime": 6889.8418,
      "train_tokens_per_second": 10334.054
    },
    {
      "epoch": 0.777884924461588,
      "grad_norm": 1.803390622138977,
      "learning_rate": 4.501940491591203e-06,
      "loss": 1.0229,
      "num_input_tokens_seen": 71313984,
      "step": 605,
      "train_runtime": 6899.7826,
      "train_tokens_per_second": 10335.686
    },
    {
      "epoch": 0.7791706846673095,
      "grad_norm": 1.9048407077789307,
      "learning_rate": 4.476067270375162e-06,
      "loss": 1.1005,
      "num_input_tokens_seen": 71454176,
      "step": 606,
      "train_runtime": 6912.9278,
      "train_tokens_per_second": 10336.312
    },
    {
      "epoch": 0.7804564448730312,
      "grad_norm": 1.7786400318145752,
      "learning_rate": 4.45019404915912e-06,
      "loss": 1.0786,
      "num_input_tokens_seen": 71608032,
      "step": 607,
      "train_runtime": 6927.2874,
      "train_tokens_per_second": 10337.096
    },
    {
      "epoch": 0.7817422050787528,
      "grad_norm": 1.9795823097229004,
      "learning_rate": 4.424320827943079e-06,
      "loss": 1.1182,
      "num_input_tokens_seen": 71703232,
      "step": 608,
      "train_runtime": 6935.3945,
      "train_tokens_per_second": 10338.739
    },
    {
      "epoch": 0.7830279652844745,
      "grad_norm": 2.045689821243286,
      "learning_rate": 4.398447606727037e-06,
      "loss": 1.0731,
      "num_input_tokens_seen": 71800608,
      "step": 609,
      "train_runtime": 6943.76,
      "train_tokens_per_second": 10340.307
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 1.9164133071899414,
      "learning_rate": 4.372574385510996e-06,
      "loss": 1.0076,
      "num_input_tokens_seen": 71902688,
      "step": 610,
      "train_runtime": 6952.5948,
      "train_tokens_per_second": 10341.849
    },
    {
      "epoch": 0.7855994856959178,
      "grad_norm": 1.7875125408172607,
      "learning_rate": 4.346701164294955e-06,
      "loss": 1.0059,
      "num_input_tokens_seen": 71999488,
      "step": 611,
      "train_runtime": 6960.8799,
      "train_tokens_per_second": 10343.446
    },
    {
      "epoch": 0.7868852459016393,
      "grad_norm": 1.7959182262420654,
      "learning_rate": 4.3208279430789134e-06,
      "loss": 1.001,
      "num_input_tokens_seen": 72127648,
      "step": 612,
      "train_runtime": 6972.1724,
      "train_tokens_per_second": 10345.075
    },
    {
      "epoch": 0.788171006107361,
      "grad_norm": 1.9085074663162231,
      "learning_rate": 4.294954721862872e-06,
      "loss": 1.0613,
      "num_input_tokens_seen": 72236512,
      "step": 613,
      "train_runtime": 6981.6576,
      "train_tokens_per_second": 10346.613
    },
    {
      "epoch": 0.7894567663130826,
      "grad_norm": 1.923269271850586,
      "learning_rate": 4.269081500646831e-06,
      "loss": 1.0021,
      "num_input_tokens_seen": 72331104,
      "step": 614,
      "train_runtime": 6989.7253,
      "train_tokens_per_second": 10348.204
    },
    {
      "epoch": 0.7907425265188043,
      "grad_norm": 1.7725319862365723,
      "learning_rate": 4.2432082794307895e-06,
      "loss": 1.0252,
      "num_input_tokens_seen": 72470080,
      "step": 615,
      "train_runtime": 7002.3646,
      "train_tokens_per_second": 10349.373
    },
    {
      "epoch": 0.7920282867245259,
      "grad_norm": 1.854164958000183,
      "learning_rate": 4.2173350582147484e-06,
      "loss": 0.9617,
      "num_input_tokens_seen": 72587776,
      "step": 616,
      "train_runtime": 7012.7177,
      "train_tokens_per_second": 10350.877
    },
    {
      "epoch": 0.7933140469302475,
      "grad_norm": 1.8561328649520874,
      "learning_rate": 4.1914618369987065e-06,
      "loss": 1.0435,
      "num_input_tokens_seen": 72709504,
      "step": 617,
      "train_runtime": 7023.2797,
      "train_tokens_per_second": 10352.642
    },
    {
      "epoch": 0.7945998071359691,
      "grad_norm": 1.8203593492507935,
      "learning_rate": 4.1655886157826655e-06,
      "loss": 1.0132,
      "num_input_tokens_seen": 72799776,
      "step": 618,
      "train_runtime": 7030.8864,
      "train_tokens_per_second": 10354.281
    },
    {
      "epoch": 0.7958855673416908,
      "grad_norm": 1.8547409772872925,
      "learning_rate": 4.1397153945666245e-06,
      "loss": 0.9842,
      "num_input_tokens_seen": 72879840,
      "step": 619,
      "train_runtime": 7037.5273,
      "train_tokens_per_second": 10355.887
    },
    {
      "epoch": 0.7971713275474124,
      "grad_norm": 1.9788426160812378,
      "learning_rate": 4.113842173350583e-06,
      "loss": 1.0385,
      "num_input_tokens_seen": 72971616,
      "step": 620,
      "train_runtime": 7045.2634,
      "train_tokens_per_second": 10357.543
    },
    {
      "epoch": 0.7984570877531341,
      "grad_norm": 1.8518086671829224,
      "learning_rate": 4.0879689521345415e-06,
      "loss": 1.0729,
      "num_input_tokens_seen": 73071392,
      "step": 621,
      "train_runtime": 7053.7889,
      "train_tokens_per_second": 10359.169
    },
    {
      "epoch": 0.7997428479588556,
      "grad_norm": 1.7593944072723389,
      "learning_rate": 4.0620957309185e-06,
      "loss": 0.9755,
      "num_input_tokens_seen": 73155936,
      "step": 622,
      "train_runtime": 7060.9396,
      "train_tokens_per_second": 10360.652
    },
    {
      "epoch": 0.8010286081645773,
      "grad_norm": 1.8963724374771118,
      "learning_rate": 4.036222509702459e-06,
      "loss": 1.0652,
      "num_input_tokens_seen": 73269600,
      "step": 623,
      "train_runtime": 7071.1619,
      "train_tokens_per_second": 10361.748
    },
    {
      "epoch": 0.8023143683702989,
      "grad_norm": 1.8380028009414673,
      "learning_rate": 4.010349288486417e-06,
      "loss": 1.0608,
      "num_input_tokens_seen": 73394752,
      "step": 624,
      "train_runtime": 7082.1878,
      "train_tokens_per_second": 10363.288
    },
    {
      "epoch": 0.8023143683702989,
      "eval_loss": 1.1012874841690063,
      "eval_runtime": 22.0115,
      "eval_samples_per_second": 45.158,
      "eval_steps_per_second": 1.454,
      "num_input_tokens_seen": 73394752,
      "step": 624
    },
    {
      "epoch": 0.8036001285760206,
      "grad_norm": 1.9703397750854492,
      "learning_rate": 3.984476067270375e-06,
      "loss": 1.0863,
      "num_input_tokens_seen": 73490912,
      "step": 625,
      "train_runtime": 7112.3423,
      "train_tokens_per_second": 10332.87
    },
    {
      "epoch": 0.8048858887817422,
      "grad_norm": 1.7578519582748413,
      "learning_rate": 3.958602846054334e-06,
      "loss": 0.9404,
      "num_input_tokens_seen": 73610944,
      "step": 626,
      "train_runtime": 7122.7514,
      "train_tokens_per_second": 10334.622
    },
    {
      "epoch": 0.8061716489874639,
      "grad_norm": 1.903511881828308,
      "learning_rate": 3.932729624838293e-06,
      "loss": 1.0024,
      "num_input_tokens_seen": 73704576,
      "step": 627,
      "train_runtime": 7130.9357,
      "train_tokens_per_second": 10335.891
    },
    {
      "epoch": 0.8074574091931854,
      "grad_norm": 1.7375211715698242,
      "learning_rate": 3.906856403622251e-06,
      "loss": 1.026,
      "num_input_tokens_seen": 73846848,
      "step": 628,
      "train_runtime": 7143.4942,
      "train_tokens_per_second": 10337.637
    },
    {
      "epoch": 0.8087431693989071,
      "grad_norm": 1.864484429359436,
      "learning_rate": 3.88098318240621e-06,
      "loss": 1.065,
      "num_input_tokens_seen": 73957984,
      "step": 629,
      "train_runtime": 7153.3216,
      "train_tokens_per_second": 10338.971
    },
    {
      "epoch": 0.8100289296046287,
      "grad_norm": 1.851105809211731,
      "learning_rate": 3.855109961190168e-06,
      "loss": 1.0817,
      "num_input_tokens_seen": 74102016,
      "step": 630,
      "train_runtime": 7166.4684,
      "train_tokens_per_second": 10340.102
    },
    {
      "epoch": 0.8113146898103504,
      "grad_norm": 2.0116357803344727,
      "learning_rate": 3.829236739974127e-06,
      "loss": 1.1515,
      "num_input_tokens_seen": 74214208,
      "step": 631,
      "train_runtime": 7192.7679,
      "train_tokens_per_second": 10317.893
    },
    {
      "epoch": 0.812600450016072,
      "grad_norm": 1.927558422088623,
      "learning_rate": 3.803363518758086e-06,
      "loss": 1.0052,
      "num_input_tokens_seen": 74299168,
      "step": 632,
      "train_runtime": 7199.9125,
      "train_tokens_per_second": 10319.454
    },
    {
      "epoch": 0.8138862102217936,
      "grad_norm": 1.961654543876648,
      "learning_rate": 3.777490297542044e-06,
      "loss": 1.0519,
      "num_input_tokens_seen": 74418176,
      "step": 633,
      "train_runtime": 7210.5863,
      "train_tokens_per_second": 10320.683
    },
    {
      "epoch": 0.8151719704275152,
      "grad_norm": 1.8367573022842407,
      "learning_rate": 3.751617076326003e-06,
      "loss": 1.011,
      "num_input_tokens_seen": 74533024,
      "step": 634,
      "train_runtime": 7220.8171,
      "train_tokens_per_second": 10321.965
    },
    {
      "epoch": 0.8164577306332369,
      "grad_norm": 1.8445461988449097,
      "learning_rate": 3.7257438551099615e-06,
      "loss": 0.9722,
      "num_input_tokens_seen": 74639200,
      "step": 635,
      "train_runtime": 7230.1938,
      "train_tokens_per_second": 10323.264
    },
    {
      "epoch": 0.8177434908389586,
      "grad_norm": 1.9773858785629272,
      "learning_rate": 3.69987063389392e-06,
      "loss": 0.9934,
      "num_input_tokens_seen": 74718656,
      "step": 636,
      "train_runtime": 7236.8882,
      "train_tokens_per_second": 10324.694
    },
    {
      "epoch": 0.8190292510446802,
      "grad_norm": 1.880348801612854,
      "learning_rate": 3.6739974126778786e-06,
      "loss": 1.0368,
      "num_input_tokens_seen": 74883680,
      "step": 637,
      "train_runtime": 7252.295,
      "train_tokens_per_second": 10325.515
    },
    {
      "epoch": 0.8203150112504018,
      "grad_norm": 1.939626932144165,
      "learning_rate": 3.648124191461837e-06,
      "loss": 1.135,
      "num_input_tokens_seen": 75006080,
      "step": 638,
      "train_runtime": 7263.2834,
      "train_tokens_per_second": 10326.746
    },
    {
      "epoch": 0.8216007714561234,
      "grad_norm": 1.7646300792694092,
      "learning_rate": 3.6222509702457957e-06,
      "loss": 1.0535,
      "num_input_tokens_seen": 75128000,
      "step": 639,
      "train_runtime": 7274.1683,
      "train_tokens_per_second": 10328.054
    },
    {
      "epoch": 0.8228865316618451,
      "grad_norm": 2.0148706436157227,
      "learning_rate": 3.5963777490297546e-06,
      "loss": 1.0519,
      "num_input_tokens_seen": 75233216,
      "step": 640,
      "train_runtime": 7283.2614,
      "train_tokens_per_second": 10329.605
    },
    {
      "epoch": 0.8241722918675667,
      "grad_norm": 1.9275726079940796,
      "learning_rate": 3.5705045278137127e-06,
      "loss": 1.0881,
      "num_input_tokens_seen": 75367840,
      "step": 641,
      "train_runtime": 7295.2124,
      "train_tokens_per_second": 10331.137
    },
    {
      "epoch": 0.8254580520732884,
      "grad_norm": 2.142881393432617,
      "learning_rate": 3.5446313065976717e-06,
      "loss": 1.2726,
      "num_input_tokens_seen": 75526528,
      "step": 642,
      "train_runtime": 7309.8561,
      "train_tokens_per_second": 10332.15
    },
    {
      "epoch": 0.82674381227901,
      "grad_norm": 1.8535045385360718,
      "learning_rate": 3.5187580853816307e-06,
      "loss": 1.0476,
      "num_input_tokens_seen": 75662336,
      "step": 643,
      "train_runtime": 7322.2115,
      "train_tokens_per_second": 10333.263
    },
    {
      "epoch": 0.8280295724847316,
      "grad_norm": 1.9154276847839355,
      "learning_rate": 3.492884864165589e-06,
      "loss": 1.131,
      "num_input_tokens_seen": 75779040,
      "step": 644,
      "train_runtime": 7332.6834,
      "train_tokens_per_second": 10334.421
    },
    {
      "epoch": 0.8293153326904532,
      "grad_norm": 1.8441919088363647,
      "learning_rate": 3.4670116429495478e-06,
      "loss": 1.049,
      "num_input_tokens_seen": 75923360,
      "step": 645,
      "train_runtime": 7346.1516,
      "train_tokens_per_second": 10335.12
    },
    {
      "epoch": 0.8306010928961749,
      "grad_norm": 2.044139862060547,
      "learning_rate": 3.441138421733506e-06,
      "loss": 1.0502,
      "num_input_tokens_seen": 76033984,
      "step": 646,
      "train_runtime": 7355.8474,
      "train_tokens_per_second": 10336.536
    },
    {
      "epoch": 0.8318868531018965,
      "grad_norm": 1.9089950323104858,
      "learning_rate": 3.415265200517465e-06,
      "loss": 1.1316,
      "num_input_tokens_seen": 76122944,
      "step": 647,
      "train_runtime": 7363.3482,
      "train_tokens_per_second": 10338.088
    },
    {
      "epoch": 0.8331726133076182,
      "grad_norm": 1.77204430103302,
      "learning_rate": 3.3893919793014234e-06,
      "loss": 1.0729,
      "num_input_tokens_seen": 76216608,
      "step": 648,
      "train_runtime": 7371.305,
      "train_tokens_per_second": 10339.636
    },
    {
      "epoch": 0.8344583735133397,
      "grad_norm": 1.8178905248641968,
      "learning_rate": 3.363518758085382e-06,
      "loss": 1.0451,
      "num_input_tokens_seen": 76310720,
      "step": 649,
      "train_runtime": 7379.274,
      "train_tokens_per_second": 10341.223
    },
    {
      "epoch": 0.8357441337190614,
      "grad_norm": 1.819828748703003,
      "learning_rate": 3.3376455368693404e-06,
      "loss": 1.0011,
      "num_input_tokens_seen": 76423424,
      "step": 650,
      "train_runtime": 7389.1813,
      "train_tokens_per_second": 10342.611
    },
    {
      "epoch": 0.837029893924783,
      "grad_norm": 1.8803701400756836,
      "learning_rate": 3.3117723156532994e-06,
      "loss": 1.0684,
      "num_input_tokens_seen": 76528768,
      "step": 651,
      "train_runtime": 7398.2156,
      "train_tokens_per_second": 10344.22
    },
    {
      "epoch": 0.8383156541305047,
      "grad_norm": 1.7649791240692139,
      "learning_rate": 3.2858990944372575e-06,
      "loss": 1.0853,
      "num_input_tokens_seen": 76637184,
      "step": 652,
      "train_runtime": 7407.6495,
      "train_tokens_per_second": 10345.682
    },
    {
      "epoch": 0.8396014143362263,
      "grad_norm": 1.8651745319366455,
      "learning_rate": 3.2600258732212165e-06,
      "loss": 1.016,
      "num_input_tokens_seen": 76752288,
      "step": 653,
      "train_runtime": 7418.1044,
      "train_tokens_per_second": 10346.617
    },
    {
      "epoch": 0.840887174541948,
      "grad_norm": 2.0024538040161133,
      "learning_rate": 3.2341526520051746e-06,
      "loss": 0.9789,
      "num_input_tokens_seen": 76872672,
      "step": 654,
      "train_runtime": 7428.7756,
      "train_tokens_per_second": 10347.96
    },
    {
      "epoch": 0.8421729347476695,
      "grad_norm": 1.891562581062317,
      "learning_rate": 3.2082794307891336e-06,
      "loss": 0.9911,
      "num_input_tokens_seen": 76990624,
      "step": 655,
      "train_runtime": 7439.4116,
      "train_tokens_per_second": 10349.021
    },
    {
      "epoch": 0.8434586949533912,
      "grad_norm": 1.8575772047042847,
      "learning_rate": 3.1824062095730925e-06,
      "loss": 1.0312,
      "num_input_tokens_seen": 77120544,
      "step": 656,
      "train_runtime": 7451.122,
      "train_tokens_per_second": 10350.192
    },
    {
      "epoch": 0.8447444551591128,
      "grad_norm": 1.7141671180725098,
      "learning_rate": 3.1565329883570506e-06,
      "loss": 0.9903,
      "num_input_tokens_seen": 77225280,
      "step": 657,
      "train_runtime": 7460.2728,
      "train_tokens_per_second": 10351.536
    },
    {
      "epoch": 0.8460302153648345,
      "grad_norm": 2.0042834281921387,
      "learning_rate": 3.130659767141009e-06,
      "loss": 1.0529,
      "num_input_tokens_seen": 77332128,
      "step": 658,
      "train_runtime": 7469.5149,
      "train_tokens_per_second": 10353.032
    },
    {
      "epoch": 0.847315975570556,
      "grad_norm": 1.9022477865219116,
      "learning_rate": 3.1047865459249677e-06,
      "loss": 1.0931,
      "num_input_tokens_seen": 77443168,
      "step": 659,
      "train_runtime": 7479.2157,
      "train_tokens_per_second": 10354.45
    },
    {
      "epoch": 0.8486017357762777,
      "grad_norm": 1.873485803604126,
      "learning_rate": 3.0789133247089263e-06,
      "loss": 1.026,
      "num_input_tokens_seen": 77537376,
      "step": 660,
      "train_runtime": 7487.2839,
      "train_tokens_per_second": 10355.875
    },
    {
      "epoch": 0.8498874959819993,
      "grad_norm": 1.8651823997497559,
      "learning_rate": 3.0530401034928852e-06,
      "loss": 1.0775,
      "num_input_tokens_seen": 77650464,
      "step": 661,
      "train_runtime": 7514.0471,
      "train_tokens_per_second": 10334.04
    },
    {
      "epoch": 0.851173256187721,
      "grad_norm": 1.8203614950180054,
      "learning_rate": 3.0271668822768433e-06,
      "loss": 1.0082,
      "num_input_tokens_seen": 77754656,
      "step": 662,
      "train_runtime": 7523.1298,
      "train_tokens_per_second": 10335.413
    },
    {
      "epoch": 0.8524590163934426,
      "grad_norm": 1.87871515750885,
      "learning_rate": 3.0012936610608023e-06,
      "loss": 1.0285,
      "num_input_tokens_seen": 77863392,
      "step": 663,
      "train_runtime": 7532.441,
      "train_tokens_per_second": 10337.073
    },
    {
      "epoch": 0.8537447765991643,
      "grad_norm": 1.8229897022247314,
      "learning_rate": 2.9754204398447613e-06,
      "loss": 1.0719,
      "num_input_tokens_seen": 77968736,
      "step": 664,
      "train_runtime": 7541.5876,
      "train_tokens_per_second": 10338.504
    },
    {
      "epoch": 0.8550305368048859,
      "grad_norm": 1.788649320602417,
      "learning_rate": 2.9495472186287194e-06,
      "loss": 1.0702,
      "num_input_tokens_seen": 78116000,
      "step": 665,
      "train_runtime": 7555.0918,
      "train_tokens_per_second": 10339.517
    },
    {
      "epoch": 0.8563162970106075,
      "grad_norm": 1.9688736200332642,
      "learning_rate": 2.9236739974126783e-06,
      "loss": 1.0683,
      "num_input_tokens_seen": 78252000,
      "step": 666,
      "train_runtime": 7567.7838,
      "train_tokens_per_second": 10340.147
    },
    {
      "epoch": 0.8576020572163292,
      "grad_norm": 1.8758140802383423,
      "learning_rate": 2.8978007761966365e-06,
      "loss": 1.0176,
      "num_input_tokens_seen": 78400896,
      "step": 667,
      "train_runtime": 7581.1157,
      "train_tokens_per_second": 10341.604
    },
    {
      "epoch": 0.8588878174220508,
      "grad_norm": 1.8528767824172974,
      "learning_rate": 2.8719275549805954e-06,
      "loss": 1.0088,
      "num_input_tokens_seen": 78535872,
      "step": 668,
      "train_runtime": 7593.4775,
      "train_tokens_per_second": 10342.544
    },
    {
      "epoch": 0.8601735776277725,
      "grad_norm": 1.8127259016036987,
      "learning_rate": 2.846054333764554e-06,
      "loss": 1.1,
      "num_input_tokens_seen": 78643040,
      "step": 669,
      "train_runtime": 7602.9127,
      "train_tokens_per_second": 10343.804
    },
    {
      "epoch": 0.861459337833494,
      "grad_norm": 1.833497405052185,
      "learning_rate": 2.8201811125485125e-06,
      "loss": 1.0125,
      "num_input_tokens_seen": 78758208,
      "step": 670,
      "train_runtime": 7612.996,
      "train_tokens_per_second": 10345.232
    },
    {
      "epoch": 0.8627450980392157,
      "grad_norm": 1.9422428607940674,
      "learning_rate": 2.794307891332471e-06,
      "loss": 1.034,
      "num_input_tokens_seen": 78856224,
      "step": 671,
      "train_runtime": 7621.3865,
      "train_tokens_per_second": 10346.703
    },
    {
      "epoch": 0.8640308582449373,
      "grad_norm": 1.9948776960372925,
      "learning_rate": 2.76843467011643e-06,
      "loss": 1.0527,
      "num_input_tokens_seen": 78963072,
      "step": 672,
      "train_runtime": 7630.5085,
      "train_tokens_per_second": 10348.337
    },
    {
      "epoch": 0.865316618450659,
      "grad_norm": 1.8143967390060425,
      "learning_rate": 2.742561448900388e-06,
      "loss": 1.0609,
      "num_input_tokens_seen": 79109120,
      "step": 673,
      "train_runtime": 7643.9846,
      "train_tokens_per_second": 10349.199
    },
    {
      "epoch": 0.8666023786563806,
      "grad_norm": 1.855102777481079,
      "learning_rate": 2.716688227684347e-06,
      "loss": 1.0352,
      "num_input_tokens_seen": 79207456,
      "step": 674,
      "train_runtime": 7652.3254,
      "train_tokens_per_second": 10350.769
    },
    {
      "epoch": 0.8678881388621023,
      "grad_norm": 2.0813863277435303,
      "learning_rate": 2.690815006468305e-06,
      "loss": 1.1286,
      "num_input_tokens_seen": 79317120,
      "step": 675,
      "train_runtime": 7661.9866,
      "train_tokens_per_second": 10352.031
    },
    {
      "epoch": 0.8691738990678238,
      "grad_norm": 1.8220653533935547,
      "learning_rate": 2.664941785252264e-06,
      "loss": 1.1715,
      "num_input_tokens_seen": 79483744,
      "step": 676,
      "train_runtime": 7677.6815,
      "train_tokens_per_second": 10352.571
    },
    {
      "epoch": 0.8704596592735455,
      "grad_norm": 1.8896037340164185,
      "learning_rate": 2.639068564036223e-06,
      "loss": 1.0194,
      "num_input_tokens_seen": 79569024,
      "step": 677,
      "train_runtime": 7684.8538,
      "train_tokens_per_second": 10354.006
    },
    {
      "epoch": 0.8717454194792671,
      "grad_norm": 1.8855335712432861,
      "learning_rate": 2.6131953428201812e-06,
      "loss": 1.0903,
      "num_input_tokens_seen": 79665344,
      "step": 678,
      "train_runtime": 7693.2234,
      "train_tokens_per_second": 10355.262
    },
    {
      "epoch": 0.8730311796849888,
      "grad_norm": 1.8641669750213623,
      "learning_rate": 2.58732212160414e-06,
      "loss": 1.0381,
      "num_input_tokens_seen": 79761120,
      "step": 679,
      "train_runtime": 7701.3015,
      "train_tokens_per_second": 10356.836
    },
    {
      "epoch": 0.8743169398907104,
      "grad_norm": 1.8233321905136108,
      "learning_rate": 2.5614489003880987e-06,
      "loss": 0.983,
      "num_input_tokens_seen": 79924640,
      "step": 680,
      "train_runtime": 7716.5881,
      "train_tokens_per_second": 10357.51
    },
    {
      "epoch": 0.875602700096432,
      "grad_norm": 1.8468879461288452,
      "learning_rate": 2.535575679172057e-06,
      "loss": 1.039,
      "num_input_tokens_seen": 80021344,
      "step": 681,
      "train_runtime": 7724.808,
      "train_tokens_per_second": 10359.007
    },
    {
      "epoch": 0.8768884603021536,
      "grad_norm": 1.7861686944961548,
      "learning_rate": 2.509702457956016e-06,
      "loss": 1.0949,
      "num_input_tokens_seen": 80140384,
      "step": 682,
      "train_runtime": 7735.3252,
      "train_tokens_per_second": 10360.312
    },
    {
      "epoch": 0.8781742205078753,
      "grad_norm": 1.9106656312942505,
      "learning_rate": 2.4838292367399743e-06,
      "loss": 1.0139,
      "num_input_tokens_seen": 80283136,
      "step": 683,
      "train_runtime": 7748.4471,
      "train_tokens_per_second": 10361.19
    },
    {
      "epoch": 0.8794599807135969,
      "grad_norm": 1.7842717170715332,
      "learning_rate": 2.457956015523933e-06,
      "loss": 1.0803,
      "num_input_tokens_seen": 80407520,
      "step": 684,
      "train_runtime": 7759.5573,
      "train_tokens_per_second": 10362.385
    },
    {
      "epoch": 0.8807457409193186,
      "grad_norm": 1.6155146360397339,
      "learning_rate": 2.4320827943078914e-06,
      "loss": 1.0904,
      "num_input_tokens_seen": 80521312,
      "step": 685,
      "train_runtime": 7769.6839,
      "train_tokens_per_second": 10363.525
    },
    {
      "epoch": 0.8820315011250401,
      "grad_norm": 1.9151482582092285,
      "learning_rate": 2.4062095730918504e-06,
      "loss": 0.9985,
      "num_input_tokens_seen": 80650112,
      "step": 686,
      "train_runtime": 7781.5312,
      "train_tokens_per_second": 10364.299
    },
    {
      "epoch": 0.8833172613307618,
      "grad_norm": 1.882094144821167,
      "learning_rate": 2.380336351875809e-06,
      "loss": 1.0523,
      "num_input_tokens_seen": 80751584,
      "step": 687,
      "train_runtime": 7790.3656,
      "train_tokens_per_second": 10365.571
    },
    {
      "epoch": 0.8846030215364834,
      "grad_norm": 1.9243850708007812,
      "learning_rate": 2.3544631306597675e-06,
      "loss": 1.0447,
      "num_input_tokens_seen": 80893472,
      "step": 688,
      "train_runtime": 7803.0693,
      "train_tokens_per_second": 10366.879
    },
    {
      "epoch": 0.8858887817422051,
      "grad_norm": 1.9028187990188599,
      "learning_rate": 2.328589909443726e-06,
      "loss": 1.124,
      "num_input_tokens_seen": 81008832,
      "step": 689,
      "train_runtime": 7813.0316,
      "train_tokens_per_second": 10368.425
    },
    {
      "epoch": 0.8871745419479267,
      "grad_norm": 1.9224538803100586,
      "learning_rate": 2.3027166882276845e-06,
      "loss": 1.0497,
      "num_input_tokens_seen": 81122272,
      "step": 690,
      "train_runtime": 7823.0001,
      "train_tokens_per_second": 10369.714
    },
    {
      "epoch": 0.8884603021536484,
      "grad_norm": 1.7827943563461304,
      "learning_rate": 2.276843467011643e-06,
      "loss": 0.9838,
      "num_input_tokens_seen": 81199616,
      "step": 691,
      "train_runtime": 7846.4547,
      "train_tokens_per_second": 10348.574
    },
    {
      "epoch": 0.8897460623593699,
      "grad_norm": 1.9258924722671509,
      "learning_rate": 2.2509702457956016e-06,
      "loss": 1.0869,
      "num_input_tokens_seen": 81317440,
      "step": 692,
      "train_runtime": 7856.7474,
      "train_tokens_per_second": 10350.013
    },
    {
      "epoch": 0.8910318225650916,
      "grad_norm": 1.7988842725753784,
      "learning_rate": 2.22509702457956e-06,
      "loss": 1.0551,
      "num_input_tokens_seen": 81392096,
      "step": 693,
      "train_runtime": 7863.0317,
      "train_tokens_per_second": 10351.236
    },
    {
      "epoch": 0.8923175827708132,
      "grad_norm": 1.9783662557601929,
      "learning_rate": 2.1992238033635187e-06,
      "loss": 1.0206,
      "num_input_tokens_seen": 81586720,
      "step": 694,
      "train_runtime": 7881.7563,
      "train_tokens_per_second": 10351.338
    },
    {
      "epoch": 0.8936033429765349,
      "grad_norm": 1.8867701292037964,
      "learning_rate": 2.1733505821474777e-06,
      "loss": 1.021,
      "num_input_tokens_seen": 81666304,
      "step": 695,
      "train_runtime": 7888.4054,
      "train_tokens_per_second": 10352.701
    },
    {
      "epoch": 0.8948891031822566,
      "grad_norm": 1.9221513271331787,
      "learning_rate": 2.147477360931436e-06,
      "loss": 1.0316,
      "num_input_tokens_seen": 81798240,
      "step": 696,
      "train_runtime": 7900.2219,
      "train_tokens_per_second": 10353.917
    },
    {
      "epoch": 0.8961748633879781,
      "grad_norm": 1.837570309638977,
      "learning_rate": 2.1216041397153947e-06,
      "loss": 1.0778,
      "num_input_tokens_seen": 81969408,
      "step": 697,
      "train_runtime": 7916.1174,
      "train_tokens_per_second": 10354.749
    },
    {
      "epoch": 0.8974606235936998,
      "grad_norm": 1.9242483377456665,
      "learning_rate": 2.0957309184993533e-06,
      "loss": 1.0736,
      "num_input_tokens_seen": 82059968,
      "step": 698,
      "train_runtime": 7923.8396,
      "train_tokens_per_second": 10356.086
    },
    {
      "epoch": 0.8987463837994214,
      "grad_norm": 1.8562523126602173,
      "learning_rate": 2.0698576972833122e-06,
      "loss": 1.0715,
      "num_input_tokens_seen": 82150944,
      "step": 699,
      "train_runtime": 7931.5687,
      "train_tokens_per_second": 10357.465
    },
    {
      "epoch": 0.9000321440051431,
      "grad_norm": 1.8766613006591797,
      "learning_rate": 2.0439844760672708e-06,
      "loss": 1.0177,
      "num_input_tokens_seen": 82278976,
      "step": 700,
      "train_runtime": 7942.9729,
      "train_tokens_per_second": 10358.713
    },
    {
      "epoch": 0.9013179042108647,
      "grad_norm": 1.8400506973266602,
      "learning_rate": 2.0181112548512293e-06,
      "loss": 0.9974,
      "num_input_tokens_seen": 82380256,
      "step": 701,
      "train_runtime": 7951.654,
      "train_tokens_per_second": 10360.141
    },
    {
      "epoch": 0.9026036644165863,
      "grad_norm": 1.869764804840088,
      "learning_rate": 1.9922380336351874e-06,
      "loss": 1.1055,
      "num_input_tokens_seen": 82467136,
      "step": 702,
      "train_runtime": 7959.0198,
      "train_tokens_per_second": 10361.469
    },
    {
      "epoch": 0.9026036644165863,
      "eval_loss": 1.1008806228637695,
      "eval_runtime": 21.9435,
      "eval_samples_per_second": 45.298,
      "eval_steps_per_second": 1.458,
      "num_input_tokens_seen": 82467136,
      "step": 702
    },
    {
      "epoch": 0.9038894246223079,
      "grad_norm": 1.8863102197647095,
      "learning_rate": 1.9663648124191464e-06,
      "loss": 1.0474,
      "num_input_tokens_seen": 82549824,
      "step": 703,
      "train_runtime": 7987.8517,
      "train_tokens_per_second": 10334.421
    },
    {
      "epoch": 0.9051751848280296,
      "grad_norm": 1.9096331596374512,
      "learning_rate": 1.940491591203105e-06,
      "loss": 0.9695,
      "num_input_tokens_seen": 82644544,
      "step": 704,
      "train_runtime": 7995.9308,
      "train_tokens_per_second": 10335.825
    },
    {
      "epoch": 0.9064609450337512,
      "grad_norm": 1.917685627937317,
      "learning_rate": 1.9146183699870635e-06,
      "loss": 1.0134,
      "num_input_tokens_seen": 82744352,
      "step": 705,
      "train_runtime": 8004.4871,
      "train_tokens_per_second": 10337.246
    },
    {
      "epoch": 0.9077467052394729,
      "grad_norm": 1.8343957662582397,
      "learning_rate": 1.888745148771022e-06,
      "loss": 1.0128,
      "num_input_tokens_seen": 82828832,
      "step": 706,
      "train_runtime": 8011.5724,
      "train_tokens_per_second": 10338.649
    },
    {
      "epoch": 0.9090324654451944,
      "grad_norm": 1.8960968255996704,
      "learning_rate": 1.8628719275549808e-06,
      "loss": 1.0537,
      "num_input_tokens_seen": 82904896,
      "step": 707,
      "train_runtime": 8017.8668,
      "train_tokens_per_second": 10340.019
    },
    {
      "epoch": 0.9103182256509161,
      "grad_norm": 1.9157270193099976,
      "learning_rate": 1.8369987063389393e-06,
      "loss": 1.0718,
      "num_input_tokens_seen": 83015168,
      "step": 708,
      "train_runtime": 8027.3151,
      "train_tokens_per_second": 10341.586
    },
    {
      "epoch": 0.9116039858566377,
      "grad_norm": 1.9855552911758423,
      "learning_rate": 1.8111254851228978e-06,
      "loss": 1.0452,
      "num_input_tokens_seen": 83164288,
      "step": 709,
      "train_runtime": 8041.0658,
      "train_tokens_per_second": 10342.446
    },
    {
      "epoch": 0.9128897460623594,
      "grad_norm": 1.8479399681091309,
      "learning_rate": 1.7852522639068564e-06,
      "loss": 1.0732,
      "num_input_tokens_seen": 83269312,
      "step": 710,
      "train_runtime": 8050.0869,
      "train_tokens_per_second": 10343.902
    },
    {
      "epoch": 0.914175506268081,
      "grad_norm": 1.8165806531906128,
      "learning_rate": 1.7593790426908153e-06,
      "loss": 1.0951,
      "num_input_tokens_seen": 83382816,
      "step": 711,
      "train_runtime": 8060.1786,
      "train_tokens_per_second": 10345.033
    },
    {
      "epoch": 0.9154612664738027,
      "grad_norm": 1.933991551399231,
      "learning_rate": 1.7335058214747739e-06,
      "loss": 1.0493,
      "num_input_tokens_seen": 83479968,
      "step": 712,
      "train_runtime": 8068.5568,
      "train_tokens_per_second": 10346.332
    },
    {
      "epoch": 0.9167470266795242,
      "grad_norm": 1.9237388372421265,
      "learning_rate": 1.7076326002587324e-06,
      "loss": 1.129,
      "num_input_tokens_seen": 83614112,
      "step": 713,
      "train_runtime": 8080.647,
      "train_tokens_per_second": 10347.453
    },
    {
      "epoch": 0.9180327868852459,
      "grad_norm": 1.7912081480026245,
      "learning_rate": 1.681759379042691e-06,
      "loss": 0.9808,
      "num_input_tokens_seen": 83697376,
      "step": 714,
      "train_runtime": 8087.6795,
      "train_tokens_per_second": 10348.75
    },
    {
      "epoch": 0.9193185470909675,
      "grad_norm": 1.951076865196228,
      "learning_rate": 1.6558861578266497e-06,
      "loss": 1.0593,
      "num_input_tokens_seen": 83815040,
      "step": 715,
      "train_runtime": 8098.1807,
      "train_tokens_per_second": 10349.86
    },
    {
      "epoch": 0.9206043072966892,
      "grad_norm": 1.97995924949646,
      "learning_rate": 1.6300129366106082e-06,
      "loss": 1.1155,
      "num_input_tokens_seen": 83897792,
      "step": 716,
      "train_runtime": 8105.0892,
      "train_tokens_per_second": 10351.248
    },
    {
      "epoch": 0.9218900675024108,
      "grad_norm": 1.942265510559082,
      "learning_rate": 1.6041397153945668e-06,
      "loss": 1.0594,
      "num_input_tokens_seen": 83994528,
      "step": 717,
      "train_runtime": 8113.4146,
      "train_tokens_per_second": 10352.55
    },
    {
      "epoch": 0.9231758277081324,
      "grad_norm": 1.8772202730178833,
      "learning_rate": 1.5782664941785253e-06,
      "loss": 1.0071,
      "num_input_tokens_seen": 84138560,
      "step": 718,
      "train_runtime": 8126.2707,
      "train_tokens_per_second": 10353.896
    },
    {
      "epoch": 0.924461587913854,
      "grad_norm": 1.8658959865570068,
      "learning_rate": 1.5523932729624839e-06,
      "loss": 1.042,
      "num_input_tokens_seen": 84260672,
      "step": 719,
      "train_runtime": 8137.107,
      "train_tokens_per_second": 10355.114
    },
    {
      "epoch": 0.9257473481195757,
      "grad_norm": 2.0019772052764893,
      "learning_rate": 1.5265200517464426e-06,
      "loss": 1.1432,
      "num_input_tokens_seen": 84396768,
      "step": 720,
      "train_runtime": 8149.6434,
      "train_tokens_per_second": 10355.885
    },
    {
      "epoch": 0.9270331083252973,
      "grad_norm": 1.8118327856063843,
      "learning_rate": 1.5006468305304011e-06,
      "loss": 1.0152,
      "num_input_tokens_seen": 84505152,
      "step": 721,
      "train_runtime": 8175.6795,
      "train_tokens_per_second": 10336.163
    },
    {
      "epoch": 0.928318868531019,
      "grad_norm": 1.8588857650756836,
      "learning_rate": 1.4747736093143597e-06,
      "loss": 1.0345,
      "num_input_tokens_seen": 84622880,
      "step": 722,
      "train_runtime": 8185.9126,
      "train_tokens_per_second": 10337.623
    },
    {
      "epoch": 0.9296046287367405,
      "grad_norm": 1.9721977710723877,
      "learning_rate": 1.4489003880983182e-06,
      "loss": 0.9971,
      "num_input_tokens_seen": 84703392,
      "step": 723,
      "train_runtime": 8192.5908,
      "train_tokens_per_second": 10339.024
    },
    {
      "epoch": 0.9308903889424622,
      "grad_norm": 1.8980185985565186,
      "learning_rate": 1.423027166882277e-06,
      "loss": 1.1035,
      "num_input_tokens_seen": 84808768,
      "step": 724,
      "train_runtime": 8201.5607,
      "train_tokens_per_second": 10340.565
    },
    {
      "epoch": 0.9321761491481839,
      "grad_norm": 1.9349688291549683,
      "learning_rate": 1.3971539456662355e-06,
      "loss": 1.0111,
      "num_input_tokens_seen": 84913856,
      "step": 725,
      "train_runtime": 8210.5039,
      "train_tokens_per_second": 10342.1
    },
    {
      "epoch": 0.9334619093539055,
      "grad_norm": 1.8501139879226685,
      "learning_rate": 1.371280724450194e-06,
      "loss": 1.0002,
      "num_input_tokens_seen": 85059104,
      "step": 726,
      "train_runtime": 8223.8201,
      "train_tokens_per_second": 10343.016
    },
    {
      "epoch": 0.9347476695596272,
      "grad_norm": 1.839609146118164,
      "learning_rate": 1.3454075032341526e-06,
      "loss": 1.0067,
      "num_input_tokens_seen": 85222624,
      "step": 727,
      "train_runtime": 8239.0013,
      "train_tokens_per_second": 10343.805
    },
    {
      "epoch": 0.9360334297653488,
      "grad_norm": 1.8927664756774902,
      "learning_rate": 1.3195342820181116e-06,
      "loss": 1.0748,
      "num_input_tokens_seen": 85403168,
      "step": 728,
      "train_runtime": 8256.2939,
      "train_tokens_per_second": 10344.008
    },
    {
      "epoch": 0.9373191899710704,
      "grad_norm": 1.9910334348678589,
      "learning_rate": 1.29366106080207e-06,
      "loss": 1.065,
      "num_input_tokens_seen": 85504928,
      "step": 729,
      "train_runtime": 8265.046,
      "train_tokens_per_second": 10345.366
    },
    {
      "epoch": 0.938604950176792,
      "grad_norm": 2.0429680347442627,
      "learning_rate": 1.2677878395860284e-06,
      "loss": 1.0859,
      "num_input_tokens_seen": 85584480,
      "step": 730,
      "train_runtime": 8271.7915,
      "train_tokens_per_second": 10346.547
    },
    {
      "epoch": 0.9398907103825137,
      "grad_norm": 1.899824619293213,
      "learning_rate": 1.2419146183699872e-06,
      "loss": 1.0421,
      "num_input_tokens_seen": 85722336,
      "step": 731,
      "train_runtime": 8284.1914,
      "train_tokens_per_second": 10347.701
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 1.7903647422790527,
      "learning_rate": 1.2160413971539457e-06,
      "loss": 0.9449,
      "num_input_tokens_seen": 85865184,
      "step": 732,
      "train_runtime": 8297.6936,
      "train_tokens_per_second": 10348.078
    },
    {
      "epoch": 0.942462230793957,
      "grad_norm": 1.8359401226043701,
      "learning_rate": 1.1901681759379045e-06,
      "loss": 1.0292,
      "num_input_tokens_seen": 85976832,
      "step": 733,
      "train_runtime": 8307.6144,
      "train_tokens_per_second": 10349.16
    },
    {
      "epoch": 0.9437479909996785,
      "grad_norm": 1.7849318981170654,
      "learning_rate": 1.164294954721863e-06,
      "loss": 0.9646,
      "num_input_tokens_seen": 86061920,
      "step": 734,
      "train_runtime": 8314.8905,
      "train_tokens_per_second": 10350.337
    },
    {
      "epoch": 0.9450337512054002,
      "grad_norm": 1.92512845993042,
      "learning_rate": 1.1384217335058215e-06,
      "loss": 0.9865,
      "num_input_tokens_seen": 86189248,
      "step": 735,
      "train_runtime": 8326.2348,
      "train_tokens_per_second": 10351.527
    },
    {
      "epoch": 0.9463195114111218,
      "grad_norm": 1.8651082515716553,
      "learning_rate": 1.11254851228978e-06,
      "loss": 1.0223,
      "num_input_tokens_seen": 86316320,
      "step": 736,
      "train_runtime": 8337.4327,
      "train_tokens_per_second": 10352.866
    },
    {
      "epoch": 0.9476052716168435,
      "grad_norm": 2.024437189102173,
      "learning_rate": 1.0866752910737388e-06,
      "loss": 1.1516,
      "num_input_tokens_seen": 86437952,
      "step": 737,
      "train_runtime": 8348.1965,
      "train_tokens_per_second": 10354.087
    },
    {
      "epoch": 0.9488910318225651,
      "grad_norm": 1.9123945236206055,
      "learning_rate": 1.0608020698576974e-06,
      "loss": 1.049,
      "num_input_tokens_seen": 86553120,
      "step": 738,
      "train_runtime": 8358.151,
      "train_tokens_per_second": 10355.534
    },
    {
      "epoch": 0.9501767920282868,
      "grad_norm": 1.8395739793777466,
      "learning_rate": 1.0349288486416561e-06,
      "loss": 1.095,
      "num_input_tokens_seen": 86637792,
      "step": 739,
      "train_runtime": 8365.3725,
      "train_tokens_per_second": 10356.717
    },
    {
      "epoch": 0.9514625522340083,
      "grad_norm": 1.8441441059112549,
      "learning_rate": 1.0090556274256147e-06,
      "loss": 1.0399,
      "num_input_tokens_seen": 86779232,
      "step": 740,
      "train_runtime": 8378.4531,
      "train_tokens_per_second": 10357.429
    },
    {
      "epoch": 0.95274831243973,
      "grad_norm": 1.9400453567504883,
      "learning_rate": 9.831824062095732e-07,
      "loss": 1.0772,
      "num_input_tokens_seen": 86871712,
      "step": 741,
      "train_runtime": 8386.2953,
      "train_tokens_per_second": 10358.771
    },
    {
      "epoch": 0.9540340726454516,
      "grad_norm": 1.7376289367675781,
      "learning_rate": 9.573091849935317e-07,
      "loss": 0.9707,
      "num_input_tokens_seen": 87009664,
      "step": 742,
      "train_runtime": 8398.846,
      "train_tokens_per_second": 10359.717
    },
    {
      "epoch": 0.9553198328511733,
      "grad_norm": 1.8850547075271606,
      "learning_rate": 9.314359637774904e-07,
      "loss": 1.0961,
      "num_input_tokens_seen": 87136544,
      "step": 743,
      "train_runtime": 8410.4197,
      "train_tokens_per_second": 10360.546
    },
    {
      "epoch": 0.9566055930568949,
      "grad_norm": 1.964850664138794,
      "learning_rate": 9.055627425614489e-07,
      "loss": 1.0545,
      "num_input_tokens_seen": 87257664,
      "step": 744,
      "train_runtime": 8421.013,
      "train_tokens_per_second": 10361.896
    },
    {
      "epoch": 0.9578913532626165,
      "grad_norm": 1.853871464729309,
      "learning_rate": 8.796895213454077e-07,
      "loss": 0.9811,
      "num_input_tokens_seen": 87335232,
      "step": 745,
      "train_runtime": 8427.4933,
      "train_tokens_per_second": 10363.133
    },
    {
      "epoch": 0.9591771134683381,
      "grad_norm": 1.8630493879318237,
      "learning_rate": 8.538163001293662e-07,
      "loss": 1.0717,
      "num_input_tokens_seen": 87423904,
      "step": 746,
      "train_runtime": 8434.9496,
      "train_tokens_per_second": 10364.484
    },
    {
      "epoch": 0.9604628736740598,
      "grad_norm": 1.965602159500122,
      "learning_rate": 8.279430789133249e-07,
      "loss": 1.0034,
      "num_input_tokens_seen": 87578368,
      "step": 747,
      "train_runtime": 8449.1667,
      "train_tokens_per_second": 10365.326
    },
    {
      "epoch": 0.9617486338797814,
      "grad_norm": 2.018232822418213,
      "learning_rate": 8.020698576972834e-07,
      "loss": 1.1624,
      "num_input_tokens_seen": 87677056,
      "step": 748,
      "train_runtime": 8457.5855,
      "train_tokens_per_second": 10366.677
    },
    {
      "epoch": 0.9630343940855031,
      "grad_norm": 1.9769020080566406,
      "learning_rate": 7.761966364812419e-07,
      "loss": 1.0854,
      "num_input_tokens_seen": 87789760,
      "step": 749,
      "train_runtime": 8467.576,
      "train_tokens_per_second": 10367.756
    },
    {
      "epoch": 0.9643201542912246,
      "grad_norm": 1.8354781866073608,
      "learning_rate": 7.503234152652006e-07,
      "loss": 1.0912,
      "num_input_tokens_seen": 87898272,
      "step": 750,
      "train_runtime": 8476.9893,
      "train_tokens_per_second": 10369.044
    },
    {
      "epoch": 0.9656059144969463,
      "grad_norm": 1.9885339736938477,
      "learning_rate": 7.244501940491591e-07,
      "loss": 1.1578,
      "num_input_tokens_seen": 88005184,
      "step": 751,
      "train_runtime": 8502.9158,
      "train_tokens_per_second": 10350.001
    },
    {
      "epoch": 0.9668916747026679,
      "grad_norm": 1.799993872642517,
      "learning_rate": 6.985769728331178e-07,
      "loss": 1.0968,
      "num_input_tokens_seen": 88145856,
      "step": 752,
      "train_runtime": 8515.7908,
      "train_tokens_per_second": 10350.871
    },
    {
      "epoch": 0.9681774349083896,
      "grad_norm": 1.7286502122879028,
      "learning_rate": 6.727037516170763e-07,
      "loss": 1.0199,
      "num_input_tokens_seen": 88253472,
      "step": 753,
      "train_runtime": 8525.1284,
      "train_tokens_per_second": 10352.157
    },
    {
      "epoch": 0.9694631951141112,
      "grad_norm": 1.8545376062393188,
      "learning_rate": 6.46830530401035e-07,
      "loss": 1.0294,
      "num_input_tokens_seen": 88410112,
      "step": 754,
      "train_runtime": 8540.0868,
      "train_tokens_per_second": 10352.367
    },
    {
      "epoch": 0.9707489553198329,
      "grad_norm": 1.7983657121658325,
      "learning_rate": 6.209573091849936e-07,
      "loss": 1.0312,
      "num_input_tokens_seen": 88517184,
      "step": 755,
      "train_runtime": 8549.4704,
      "train_tokens_per_second": 10353.528
    },
    {
      "epoch": 0.9720347155255545,
      "grad_norm": 1.9300833940505981,
      "learning_rate": 5.950840879689522e-07,
      "loss": 0.9773,
      "num_input_tokens_seen": 88644960,
      "step": 756,
      "train_runtime": 8561.4036,
      "train_tokens_per_second": 10354.022
    },
    {
      "epoch": 0.9733204757312761,
      "grad_norm": 1.8282893896102905,
      "learning_rate": 5.692108667529108e-07,
      "loss": 0.9824,
      "num_input_tokens_seen": 88763680,
      "step": 757,
      "train_runtime": 8572.4426,
      "train_tokens_per_second": 10354.538
    },
    {
      "epoch": 0.9746062359369978,
      "grad_norm": 2.0415329933166504,
      "learning_rate": 5.433376455368694e-07,
      "loss": 1.0697,
      "num_input_tokens_seen": 88915392,
      "step": 758,
      "train_runtime": 8586.3403,
      "train_tokens_per_second": 10355.447
    },
    {
      "epoch": 0.9758919961427194,
      "grad_norm": 1.8638561964035034,
      "learning_rate": 5.174644243208281e-07,
      "loss": 1.1258,
      "num_input_tokens_seen": 89018304,
      "step": 759,
      "train_runtime": 8595.2138,
      "train_tokens_per_second": 10356.729
    },
    {
      "epoch": 0.9771777563484411,
      "grad_norm": 1.989698886871338,
      "learning_rate": 4.915912031047866e-07,
      "loss": 1.0544,
      "num_input_tokens_seen": 89134688,
      "step": 760,
      "train_runtime": 8605.465,
      "train_tokens_per_second": 10357.916
    },
    {
      "epoch": 0.9784635165541626,
      "grad_norm": 1.9920203685760498,
      "learning_rate": 4.657179818887452e-07,
      "loss": 0.9777,
      "num_input_tokens_seen": 89292992,
      "step": 761,
      "train_runtime": 8620.2862,
      "train_tokens_per_second": 10358.472
    },
    {
      "epoch": 0.9797492767598843,
      "grad_norm": 1.792285680770874,
      "learning_rate": 4.3984476067270383e-07,
      "loss": 1.011,
      "num_input_tokens_seen": 89397312,
      "step": 762,
      "train_runtime": 8629.3843,
      "train_tokens_per_second": 10359.64
    },
    {
      "epoch": 0.9810350369656059,
      "grad_norm": 1.7668837308883667,
      "learning_rate": 4.139715394566624e-07,
      "loss": 1.0001,
      "num_input_tokens_seen": 89473952,
      "step": 763,
      "train_runtime": 8635.848,
      "train_tokens_per_second": 10360.76
    },
    {
      "epoch": 0.9823207971713276,
      "grad_norm": 2.0289688110351562,
      "learning_rate": 3.8809831824062096e-07,
      "loss": 1.0785,
      "num_input_tokens_seen": 89572416,
      "step": 764,
      "train_runtime": 8644.254,
      "train_tokens_per_second": 10362.076
    },
    {
      "epoch": 0.9836065573770492,
      "grad_norm": 1.9272173643112183,
      "learning_rate": 3.6222509702457956e-07,
      "loss": 1.0206,
      "num_input_tokens_seen": 89671456,
      "step": 765,
      "train_runtime": 8652.7996,
      "train_tokens_per_second": 10363.288
    },
    {
      "epoch": 0.9848923175827708,
      "grad_norm": 2.0589964389801025,
      "learning_rate": 3.3635187580853815e-07,
      "loss": 1.029,
      "num_input_tokens_seen": 89745216,
      "step": 766,
      "train_runtime": 8659.0809,
      "train_tokens_per_second": 10364.289
    },
    {
      "epoch": 0.9861780777884924,
      "grad_norm": 1.846987247467041,
      "learning_rate": 3.104786545924968e-07,
      "loss": 1.0234,
      "num_input_tokens_seen": 89840896,
      "step": 767,
      "train_runtime": 8667.3147,
      "train_tokens_per_second": 10365.482
    },
    {
      "epoch": 0.9874638379942141,
      "grad_norm": 1.9260659217834473,
      "learning_rate": 2.846054333764554e-07,
      "loss": 1.2661,
      "num_input_tokens_seen": 89958592,
      "step": 768,
      "train_runtime": 8677.7751,
      "train_tokens_per_second": 10366.55
    },
    {
      "epoch": 0.9887495981999357,
      "grad_norm": 1.8034597635269165,
      "learning_rate": 2.5873221216041403e-07,
      "loss": 1.016,
      "num_input_tokens_seen": 90066944,
      "step": 769,
      "train_runtime": 8687.2356,
      "train_tokens_per_second": 10367.734
    },
    {
      "epoch": 0.9900353584056574,
      "grad_norm": 1.848668098449707,
      "learning_rate": 2.328589909443726e-07,
      "loss": 1.1509,
      "num_input_tokens_seen": 90199456,
      "step": 770,
      "train_runtime": 8699.3479,
      "train_tokens_per_second": 10368.531
    },
    {
      "epoch": 0.991321118611379,
      "grad_norm": 1.843025803565979,
      "learning_rate": 2.069857697283312e-07,
      "loss": 1.4619,
      "num_input_tokens_seen": 90314368,
      "step": 771,
      "train_runtime": 8709.3929,
      "train_tokens_per_second": 10369.766
    },
    {
      "epoch": 0.9926068788171006,
      "grad_norm": 1.9316155910491943,
      "learning_rate": 1.8111254851228978e-07,
      "loss": 1.1308,
      "num_input_tokens_seen": 90455488,
      "step": 772,
      "train_runtime": 8722.3601,
      "train_tokens_per_second": 10370.529
    },
    {
      "epoch": 0.9938926390228222,
      "grad_norm": 1.8842840194702148,
      "learning_rate": 1.552393272962484e-07,
      "loss": 1.181,
      "num_input_tokens_seen": 90566848,
      "step": 773,
      "train_runtime": 8732.2133,
      "train_tokens_per_second": 10371.58
    },
    {
      "epoch": 0.9951783992285439,
      "grad_norm": 1.918171763420105,
      "learning_rate": 1.2936610608020701e-07,
      "loss": 1.0189,
      "num_input_tokens_seen": 90670336,
      "step": 774,
      "train_runtime": 8741.1414,
      "train_tokens_per_second": 10372.826
    },
    {
      "epoch": 0.9964641594342655,
      "grad_norm": 2.0272796154022217,
      "learning_rate": 1.034928848641656e-07,
      "loss": 1.0934,
      "num_input_tokens_seen": 90764096,
      "step": 775,
      "train_runtime": 8749.2474,
      "train_tokens_per_second": 10373.932
    },
    {
      "epoch": 0.9977499196399872,
      "grad_norm": 1.906491994857788,
      "learning_rate": 7.76196636481242e-08,
      "loss": 1.0435,
      "num_input_tokens_seen": 90880416,
      "step": 776,
      "train_runtime": 8759.5236,
      "train_tokens_per_second": 10375.041
    },
    {
      "epoch": 0.9990356798457087,
      "grad_norm": 1.8904526233673096,
      "learning_rate": 5.17464424320828e-08,
      "loss": 1.0791,
      "num_input_tokens_seen": 91001792,
      "step": 777,
      "train_runtime": 8770.7656,
      "train_tokens_per_second": 10375.581
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.2318460941314697,
      "learning_rate": 2.58732212160414e-08,
      "loss": 0.9926,
      "num_input_tokens_seen": 91103430,
      "step": 778,
      "train_runtime": 8781.8231,
      "train_tokens_per_second": 10374.091
    }
  ],
  "logging_steps": 1,
  "max_steps": 778,
  "num_input_tokens_seen": 91103430,
  "num_train_epochs": 1,
  "save_steps": 30,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5167484420599808e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}